BR112014008684B1 - METHODS OF PRODUCTION OF HETEROMULTIMERIC PROTEINS AND METHOD OF PRODUCTION OF BISPECIFIC ANTIBODY - Google Patents

Abstract

Heteromultimeric protein production method and bispecific antibody production method are described herein as methods of efficiently producing a heteromultimeric protein such as a bispecific antibody. Heteromultimeric proteins may be capable of specifically binding to more than one target molecule or different epitopes on a single target molecule. the methods modulate parameters to enhance the assembly of heteromultimeric proteins in higher throughput and efficiency than would otherwise be possible. Compositions comprising a hinge-containing polypeptide, such as a half-antibody, are also described.

Description

[001] The present invention refers to and claims the priority benefit of the US Provisional Patent Applications with serial numbers 61/545863, filed October 11, 2011, 61/546503, filed October 12, 2011 , 61/560704, filed November 16, 2011, and 61/676837, filed July 27, 2012. The descriptive report of each of the Provisional Orders referenced above is incorporated in its entirety by reference herein.

FIELD OF INVENTION

[002] This descriptive report refers to improved compositions and methods of assembling heteromultimeric proteins such as bispecific antibodies.

BACKGROUND OF THE INVENTION

[003] IgG-like monoclonal antibodies contain two identical antigen-binding arms and a constant domain (Fc). Antibodies with different specificity in their binding arms do not normally occur in nature and therefore need to be elaborated with the aid of chemical engineering (such as chemical cross-linking etc.), recombinant DNA and/or cell fusion technology.

[004] Bispecific antibodies can simultaneously bind two different antigens. This property allows the development of therapeutic strategies that are not possible with conventional monoclonal antibodies. The large picture of imaginative bispecific antibody formats that have been developed reflects the strong interest in these molecules. See Berg, J., Lotscher, E., Steimer, K.S. et al, Bispecific Antibodies that Mediate Killing of Cells Infected with Human Immunodeficiency Virus of Any Strain, Proc. Natl. Academic Sci. U.S.A. 1991) 88 (11): 4723-4727 and Fischer, N. and Leger, O., Biospecific Antibodies: Molecules that Enable Novel Therapeutic Strategies, Pathobiology (2007) 74: 3-14.

[005] Another class of multispecific molecules are recombinant fusion proteins. Recombinant fusion proteins consisting of the extracellular domain of immunoregulatory proteins and the constant domain (Fc) of immunoglobulin (Ig) represent a growing class of human therapeutics. Immunoadhesins combine the binding region of a protein sequence, with desired specificity, with the effector domain of an antibody. Immunoadhesins have two important properties that are significant for their potential as therapeutic agents: target specificity and pharmacokinetic stability (in vivo half-life that is comparable to that of antibodies). Immunoadhesins can be used as antagonists to inhibit or block harmful interactions or as agonists to mimic or amplify physiological reactions. See Chamow, S.M., Zhang, D.Z, Tan, X.Y. et al, A Humanized, Bispecific Immunoadhesin-Antibody that Retargets CD3+ Effectors to Kill HIV-1 Infected Cells, J. Hematother. 1995; 4(5): 439-446.

[006] Other multispecific molecules have been discussed in other publications. Examples include, but are not limited to: Fisher et al, Pathobiology (2007) 74: 3-14 (analysis of several bispecific formats); U.S. Patent No. 6,660,843, issued December 9, 2003 to Feige et al (peptide bodies); U.S. Patent Published No. 2002-004587, issued January 10, 2002 (multispecific antibodies); U.S. Patent No. 7,612,181, issued November 3, 2009 to Wu et al (double variable domain format); U.S. Patent No. 6,534,628, Nord, K. et al, Prot. Eng. (1995) 8:601-608, Nord, K. et al, Nat. Biotech. (1997) 15: 772-777 and Gronwall et al, Biotechnol. Appl. Biochem. (2008), June; 50 (Pt 2): 97-112 (affinity bodies); Martens et al, Clin. Cancer Res. (2006), 12: 6144-6152 and Jin et al, Cancer Res. (2008) 68 (11): 4360-4368 (one-armed antibodies); Bostrom et al, Science (2009) 323: 1610-1614 (Dual-acting Fab, also known as mixed valence antibodies). Other formats are known to those skilled in the art.

[007] The manufacture of clinical grade material remains a challenge for antibodies in general and especially for the multispecific molecules described above. As indicated above, there are many processes for producing molecules with mixed binding arms, that is, binding arms that are not identical with each other. Each of these methods, however, has its drawbacks.

[008] The chemical screening is labor intensive, as the relevant substance may still need to be purified from homodimers and other undesirable by-products. Furthermore, chemical modification steps can alter the integrity of proteins, in order to generate low stability. Therefore, this method is often inefficient and can lead to loss of antibody activity.

[009] Cell fusion technology (such as hybrid hybridomas) express two randomly assembled light and two heavy chains, leading to the generation of ten antibody combinations. The desired heteromultimeric antibodies are only a small fraction of the antibodies produced in this way. Purification of the desired heteromultimeric proteins dramatically reduces production yields and increases manufacturing costs.

[010] Recombinant DNA methods have been used to generate several heteromultimeric formats, such as single-chain Fv, diabodies, etc., which do not comprise an Fc domain. A major disadvantage of this type of antibody molecule is the lack of the Fc domain and therefore the antibody's ability to trigger an effector function (such as complement activation, Fc receptor binding etc.). In this way, a bispecific antibody that comprises a functional Fc domain is desired.

[011] Recombinant DNA methods have also been used to generate bispecific “knob-in-hole” antibodies. See U.S. Patent Application No. 20030078385 (Arathoon et al - Genentech). A restriction of this strategy is the fact that the light chains of the two parental antibodies need to be identical to avoid mismatch and the formation of inactive and/or unwanted molecules when expressed in the same cell.

[012] Furthermore, one of the limiting events during combination and purification is the redox efficiency. Oxidized heterodimer typically makes up only 70-80% of the protein after this step (BioAnalyzer and MS-TOF). The remaining 20-30% antibody is dimeric and non-covalently bound (SEC-LLS). This can be removed, but it has a significant impact on overall earnings. Thus, there remains a need to increase the overall yield of antibody production, especially heterodimers. Methods that can increase the overall yield of bispecific antibodies, heterodimers, and the like are described herein. These and other aspects and advantages of the present invention will be apparent from the description of the invention provided herein.

BRIEF DESCRIPTION OF THE INVENTION

[013] The production of heteromultimeric proteins assembled from two or more polypeptides that contain joints, such as multispecific antibodies from two or more half-antibodies, using current methods, presents disadvantages, which include the manufacture of a mixture of products, reduction efficiency and reduction/elimination of the effector function, among others. Furthermore, aggregation and precipitation often occur during the preparation of each hinge-containing polypeptide and during the assembly or combination of the heteromultimers. Aggregation and precipitation can greatly reduce the yield of the desired heteromultimer. It is therefore desirable to produce heteromultimeric proteins more efficiently and at higher levels.

[014] Efficient production methods/processes for the economical production of heteromultimeric proteins, such as multispecific antibodies, using or modulating one or more of the following are described herein, including, without limitation: a stabilizer, a solubilizer, a reduction condition , selected pH, selected temperature, etc. The methods according to the present invention described herein reduced protein loss in preparation and/or aggregation and increased the overall yield of heteromultimeric protein production, such as bispecific antibody production.

[015] In one aspect, a method of producing heteromultimeric protein is provided, wherein said method comprises: a. providing a first hinge-containing polypeptide under pH 4-9, preferably 5-9, in the presence of a first solubilizer, wherein the first hinge-containing polypeptide comprises a heteromultimerization domain; B. providing a second hinge-containing polypeptide under pH 4-9, preferably 5-9, in the presence of a second solubilizer, wherein the second hinge-containing polypeptide comprises a heteromultimerization domain; ç. mixing the first and second articulating-containing polypeptides in reducing condition to form an assembly mixture; and d. incubating the assembly mixture to form or produce a heteromultimeric protein comprising the first and second hinge-containing polypeptides, wherein the first hinge-containing polypeptide interacts with the second hinge-containing polypeptide in the heteromultimerization domain.

[016] In certain embodiments of this aspect, step a and/or step b is preceded by the step of purifying the first and/or second hinge-containing polypeptide. In certain specific embodiments, the first and/or second hinge-containing polypeptide is purified by Protein A.

[017] In another aspect, a method of forming or producing bispecific antibodies is provided, wherein said method comprises: a. providing a first half-antibody under pH 4-9, preferably 5-9, in the presence of a first solubilizer, wherein the first half-antibody comprises a heteromultimerization domain; B. providing a second half-antibody under pH 4-9, preferably 5-9, in the presence of a second solubilizer, wherein the second half-antibody comprises a heteromultimerization domain; ç. mixing the first and second half-antibodies in a reducing condition to form an assembly mixture; and d. incubating the assembly mixture to form or produce a bispecific antibody that comprises the first and second half-antibody, wherein the first half-antibody interacts with the second half-antibody in the heteromultimerization domain.

[018] In certain embodiments of this aspect, step a and/or step b is preceded by the step of purifying the first and/or second half-antibody. In certain specific embodiments, the first and/or second half-antibody is purified by Protein A.

[019] In a further aspect, a method of producing heteromultimers is provided, wherein said method comprises providing a mixture containing arginine of hinge-containing polypeptides, wherein said mixture has a pH of 4 to 9, preferably 5-9, addition of a weak reductant and incubation under heteromultimer production conditions.

[020] In yet another aspect, a method of producing heteromultimeric protein is provided, wherein said method comprises: a. obtaining a first protein A-purified linkage-containing polypeptide; B. obtaining a second protein A-purified hinge-containing polypeptide; ç. adjusting the pH of each half-antibody from 4 to 9; d. mixing the first and second articulation-containing polypeptides to obtain an assembly mixture; and. adding a molar excess of a weak reducer to the assembly mixture; and f. incubating the assembly mixture to form a heteromultimeric protein that comprises the first and second hinge-containing polypeptides.

[021] In another aspect, a method of producing heteromultimeric protein is provided, wherein said method comprises: a. obtaining a first protein A-purified linkage-containing polypeptide; B. obtaining a second protein A-purified hinge-containing polypeptide; ç. adjusting the pH of each hinge-containing polypeptide by 4 to 9 in the presence of L-arginine; d. mixing the first and second hinge-containing polypeptides to obtain a mixed hinge-containing polypeptide pool; and is. incubation to form a heteromultimeric protein comprising the first and second hinge-containing polypeptides.

[022] In certain embodiments of this aspect, the pooled polypeptide-containing scrambled joint is incubated in a reducing condition. In certain embodiments, the hinge-containing polypeptide comprises a half-antibody, immunoadhesin or functional fragment thereof. In some other embodiments, arginine is present at a concentration of 20 mM to 1 M, 20 mM to 200 mM, or 50 mM to 200 mM. In some other embodiments, PVP is added to step d or step e. In certain embodiments, the pH is adjusted after mixing.

[023] The depositors of the present unexpectedly discovered that the maintenance of intermediate pH of a hinge-containing polypeptide such as half-antibody can promote the conformational change that improved the subsequent assembly of the hinge-containing polypeptides. In certain embodiments, the intermediate pH is from pH 4 to 9, preferably 5 to 9 or at least pH 5, at least pH 5.5, at least pH 5.7, pH greater than 5, pH greater than 5.5, pH greater than 5.7, 5 to 9, 5 to 8, 5.5 to 8, 5.5 to 9, 5.7 to 8, 5.7 to 9, 6 to 8, 6 to 9, 7 to 8, 7.5 to 8.5 or 7 to 8.5. A solubilizer can be added to prevent or minimize pH-induced precipitation of the hinge-containing polypeptide. In certain specific embodiments, the solubilizer is added prior to maintaining the intermediate pH. In certain embodiments, the first solubilizer and the second solubilizer are selected from the group consisting of arginine, histidine and sucrose, preferably arginine and/or histidine. In some other embodiments, arginine is a salt of arginine and/or histidine is a salt of histidine. In some other embodiments, arginine is a derivative of arginine and/or histidine is a derivative of histidine. In some other embodiments, the arginine or histidine is L-arginine or L-histidine. In some other embodiments, the arginine or histidine is arginine HCl or histidine HCl. In some other embodiments, the arginine or histidine is arginine phosphate or histidine phosphate. In certain embodiments, the first and second solubilizers are different; while, in other embodiments, the first and second solubilizers are identical. In certain preferred embodiments, the first solubilizer and the second solubilizer comprise arginine. In still other embodiments, arginine is present at a concentration of 20 mM to 1 mM, 20 mM to less than 1 mM, 20 mM to 100 mM, 20 mM to 200 mM, 20 mM to 300 mM, 20 mM to 400 mM , 50 mM to 100 mM, 50 mM to 150 mM, 50 mM to 200 mM, 50 mM to 250 mM or 50 mM to 300 mM, preferably from 20 mM to 200 mM. In still other embodiments, the solubilizer comprises an arginine derivative which includes, without limitation, acetylarginine. In other embodiments, the first solubilizer and the second solubilizer comprise histidine present at a concentration of 20 mM to 1 M, 20 mM to less than 1 M, 20 mM to less than 500 mM, 20 mM to 100 mM, 20 mM to 200 mM, 20mM to 300mM, 20mM to 400mM, 50mM to 100mM, 50mM to 150mM, 50mM to 200mM, 50mM to 250mM, 50mM to 300mM, 50mM to 400mM, 50 mM to 500 mM or 50 mM to 600 mM. In certain preferred embodiments, the solubilizer is added at a concentration of 50 mM. In certain preferred embodiments, the solubilizer is added at a concentration of 200 mM. In certain specific embodiments, arginine or histidine is added at a concentration of 20 mM, 50 mM, 100 mM or 200 mM. In some other specific embodiments, the first and/or second hinge-containing polypeptides are provided in the presence of arginine and histidine. In other embodiments, arginine and histidine are present at a concentration of 20 mM to 1 mM, 20 mM to less than 1 mM, 20 mM to 100 mM, 20 mM to 200 mM, 20 mM to 300 mM, 20 mM to 400mM, 50mM to 100mM, 50mM to 150mM, 50mM to 200mM, 50mM to 250mM or 50mM to 300mM, preferably from 50mM to 200mM each.

[024] In certain embodiments, the first and second hinge-containing polypeptides are mixed prior to maintaining intermediate pH (ie, pH adjustment). In some other embodiments, the first and second hinge-containing polypeptides are mixed after adjusting the pH on the first hinge-containing polypeptide and the second hinge-containing polypeptide separately. In certain embodiments, a solubilizer is added prior to pH adjustment.

[025] In certain embodiments, the first hinge-containing polypeptide and the second hinge-containing polypeptide are separately purified prior to mixing; while, in other embodiments, the first hinge-containing polypeptide and the second hinge-containing polypeptide are co-purified after mixing. In certain specific embodiments, the hinge-containing polypeptide comprises a half-antibody. In certain embodiments, the assembled heteromultimeric protein can be subjected to further purification.

[026] Any suitable method can be used for purification, including, without limitation, purification by protein A chromatography, protein G chromatography, hydrophobic interaction chromatography (HIC), fractionation on immunoaffinity column, ethanol precipitation, chromatography reverse phase on silica or on an ion exchange resin such as DEAE, chromatoconcentration, SDS-PAGE, ammonium sulfate precipitation and gel filtration using, for example, Sephadex G-75 and other similar purification methods, as well as their combinations.

[027] In certain embodiments, the hinge-containing polypeptide such as half-antibody is purified by means of Protein A or Protein G chromatography. In another embodiment, the first and second hinge-containing polypeptides are mixed prior to purification with protein A and co-purified over protein A. In certain embodiments, the pH is adjusted after mixing the Protein A purified polypeptides. In other embodiments, the pH is adjusted before mixing the Protein A purified polypeptides. solubilizer before pH adjustment.

[028] In some other embodiments, the hinge-containing polypeptide is purified by means of HIC or an ion exchange column. It is within the skill of those skilled in the art to select appropriate purification methods. The hinge-containing polypeptide can be purified, for example, by a Protein A column followed by an ion exchange column; the hinge-containing polypeptide can also be purified by a Protein A column followed by a gel filtration column and/or HIC. In other examples, the hinge-containing polypeptide can be purified by one or more ion exchange columns prior to purification by a Protein A column. purification of the hinge-containing polypeptides do not contain arginine and/or histidine.

[029] In still other embodiments, the half-antibody eluted from the Protein A matrix or other column matrix under acidic pH is adjusted to intermediate pH. This subsequent pH adjustment (also called maintenance of intermediate pH) can cause precipitation of the hinge-containing polypeptide such as half-antibody and, therefore, generate reduced yield of assembled heteromultimeric protein. Thus, in certain embodiments, the half-antibody eluted from the Protein A or Protein G column under acidic pH is provided in the presence of a solubilizer prior to pH adjustment. If a pH adjustment step is not required, in certain embodiments, a solubilizer is preferably added to the purified hinge-containing polypeptide to prevent or reduce precipitation and/or aggregation.

[030] In addition to maintaining an intermediate pH, depositors of the present have unexpectedly discovered that heating can increase the conformational and/or assembly alteration of hinge-containing polypeptides such as half-antibodies. Accordingly, in certain embodiments, one, more or all steps ad of the methods according to the present invention are heated at a temperature of 15°C to 39°C, 15°C to 42°C, 18°C to 37°C , 20 °C to 42 °C, 20 °C to 25 °C, 25 °C to 42 °C, 25 °C to 35 °C, 25 °C to 39 °C, 30 °C to 35 °C, 32 °C to 35 °C or 32 °C to 37 °C, preferably 35 °C to 37 °C, for at least thirty minutes. In certain embodiments, the incubation time is up to 72 hours, especially at room temperature. In some embodiments, the incubation time is three hours at 35 °C. In some other embodiments, the temperature is at or about 30°C, 35°C, or 37°C.

[031] Warming, however, can also increase aggregation and/or precipitation. Accordingly, in certain specific embodiments, a solubilizer is added to the antibody medium which has eluted from a Protein A or Protein G column prior to heating.

[032] In certain embodiments, the hinge-containing polypeptide comprises a half-antibody, immunoadhesin or functional fragment thereof. In certain specific embodiments, the hinge-containing polypeptide comprises an Fc component.

[033] In certain specific embodiments, the first and/or second hinge-containing polypeptides comprise a half-antibody. In certain embodiments, the half-antibody is an IgG half-antibody. In certain specific embodiments, the IgG half-antibody is of the IgG1, IgG4 or IgG2 isotype. In certain advantageous embodiments, the signal peptide of the immunoglobulin molecule is retained to facilitate secretion of the half-antibody, especially when produced in mammalian cells. In certain embodiments, the method according to the present invention comprises providing first and second half-antibodies under pH 5-9 in the presence of arginine at a concentration of about 50 mM and alternatively or additionally histidine at a concentration of about 200 mM. In certain embodiments, the first and/or second half-antibodies comprise an antigen-binding domain specific for a different antigen each or a different epitope on the same antigen and the assembled complete antibody is a bispecific antibody. In some other embodiments, the first and second half-antibodies are of the same isotype, while, in other embodiments, the first and second half-antibodies are of different isotypes.

[034] The half-antibody may comprise a VL domain, VH domain, hinge domain, CH2 domain and/or CH3 domain. The half-antibody can also be a single-chain polypeptide which further comprises a teter and said single-chain polypeptide comprises domains positioned between them in the N-terminal to C-terminal direction as follows: VL -teter-VH-hinge-CH2 -CH3. In some other embodiments, the half-antibody further comprises a CL domain and a CH1 domain; and, in further embodiments, the half-antibody may be a single-chain polypeptide which further comprises a tether and said single-chain polypeptide comprises domains positioned from each other in the N-terminal to C-terminal direction as follows: VL-CL- tether-VH-CH1-joint-CH2-CH3.

[035] The tether can comprise one or more residues of Glycine (G) and Serine (S). In other embodiments, the teter comprises GGS repeats. The teter, for example, is 15 to 50 amino acids long. In a specific embodiment, the teter is 20 to 32 amino acids long, such as 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 or 32 amino acids long. In certain embodiments, the teter is divisible. In other embodiments, the teter may or may not be divisible from the protein. In certain preferred embodiments, the teter is divisible into two locations at or near the N- and C-terminus of the teter by the same enzyme. In one embodiment, the tether comprises the cleavage site for proteases such as furin. In a further embodiment, the teter is cleaved by furin at the RXRXRR cleavage site (SEQ ID NO: 1), where X is any amino acid. In some embodiments, the first hinge-containing polypeptide is a half-antibody and the second hinge-containing polypeptide is a half-single-chain antibody.

[036] In another embodiment, the present invention provides a protein comprising a tether and a complex of Fc components, wherein the teter may or may not be divisible from the protein.

[037] In further embodiments, the first and second hinge-containing polypeptides comprise a heteromultimerization domain. The heteromultimerization domain can be a knob-in-hole mutation, leucine zipper, electrostatic and the like. The first hinge-containing polypeptide may comprise a knob and the second hinge-containing polypeptide may comprise a hole. In certain embodiments, the hinge-containing polypeptide comprises a half-antibody, the first half-antibody comprises a bulge (knob), and the second half-antibody comprises a hole (hole).

[038] In some embodiments, the method comprises adding arginine to a final concentration of 20 mM to 1 M before pH adjustment. In some embodiments, arginine is added to a final concentration of 50 mM to 600 mM. In some embodiments, arginine is added to a final concentration of 50 mM to 100 mM.

[039] In certain embodiments, the method includes incubating each set of Protein A and hinge-containing polypeptide under pH 5 to 8 prior to mixing the first and second half-antibodies. In other embodiments, Protein A pools are mixed and then the pH is adjusted to 5 to 8.

[040] In some embodiments, the methods described herein comprise incubating the pool of polypeptide containing mixed hinge or mixed antibody medium at a temperature of 15 °C to 39 °C, preferably from 18 °C to 37 °C, from more preferably from 20°C to 25°C and more preferably from 32°C to 37°C for at least thirty minutes.

[041] Joint-containing polypeptides can be produced, for example, by a bacterial cell, a yeast cell, a bacillovirus in an insect cell or a mammalian cell. In certain embodiments, the hinge-containing polypeptide is produced by a bacterial cell, particularly E. coli. In some other specific embodiments, the hinge-containing polypeptide is produced by a mammalian cell, particularly a CHO cell. In certain specific embodiments, the hinge-containing polypeptide comprises a half-antibody.

[042] Linkage-containing polypeptides can interact to form a dimer or multimer via the heterodimerization domain. In certain embodiments, the interaction between the first and second hinge-containing polypeptides at the interface of the heterodimerization domains is a knob-in-hole interaction, hydrophobic interaction, and/or electrostatic interaction. In some other embodiments, the heteromultimerization domain comprises one or more of a bulge (such as a bulge-knob), orifice (such as a hollow-hole), leucine zipper, coiled coil, or coiled coil residue. polar amino acid capable of forming an electrostatic interaction or combinations thereof. In certain embodiments, the first hinge-containing polypeptide comprises a knob and the second hinge-containing polypeptide comprises a hole (hole). In some other embodiments, the interaction involves a hydrophobic interaction and an electrostatic interaction. In certain embodiment examples, the heteromultimerization domain of each of the first and second hinge-containing polypeptides comprises a bulge (knob) or hole (hole) and an amino acid residue capable of forming an electrostatic interaction. Those skilled in the art understand that the heterodimerization domain can comprise more than one form of interaction, such as hydrophobic interaction and bulge (knob) and hole, bulge (knob) and hole (hole) and leucine closure (leucine) zipper) etc. In certain embodiments, the hinge-containing polypeptide further comprises a tether. In certain specific embodiments, the hinge-containing polypeptide comprises a half-antibody.

[043] In certain specific embodiments, the assembly mixture is present in a reducing condition, preferably a weak reducing condition, which has oxidation potential of -50 to -600 mV, -100 to -600 mV, -200 to -600 mV, -100 to -500 mV, -150 to -300 mV, more preferably -300 to -500 mV, preferably greater than about -400 mV, under conditions that promote the assembly of the heteromultimeric proteins such as when the pH is from 7 to 9 and the temperature is from 15 °C to 39 °C. In certain embodiments, a reducer is added to step c or step d to prepare a desired reducing condition during assembly. In some other embodiments, the reductant is selected from the group consisting of dithiothreitol (DDTT), tris(2-carboxyethyl)phosphine (TCEP), thioglycolic acid, ascorbic acid, thiolacetic acid, glutathione (GSH), betamercaptoethylamine, cysteine /cystine, glutathione (GSH), cysteamine/cystamine, glycylcysteine and betamercaptoethanol, preferably GSH. In certain preferred embodiments, the reductant, preferably a weak reductant, is selected from the group consisting of glutathione (GSH), betamercaptoethylamine, cysteine/cystine, glutathione (GSH)/glutathione disulfide (GSSG), cysteamine/cystamine, glycylcysteine and betamercaptoethanol, preferably GSH. In certain embodiments, the reducer is not DTT.

[044] In some other embodiments, the reductant is added to the reaction mixture in molar excess of 2-600X, 2-200X, 2-300X, 2-400X, 2500X, 2-20X, 2-8X, 20-50X, 50-600X, 50-200X or 100-300X, preferably 50-400X, more preferably 100-300X and most preferably 200X with respect to the total amount of hinge-containing polypeptides. In certain embodiments, the assembly mixture has a pH of 7 to 9, preferably pH 8.5. In certain specific embodiments, the hinge-containing polypeptide is a half-antibody.

[045] In some embodiments, the reductant is added to the first and second hinge-containing polypeptides prior to mixing. Preferably, the addition is less than one hour, more preferably less than fifteen minutes, and most preferably less than five minutes prior to blending.

[046] In certain embodiments, the method further comprises adding a stabilizer to the reaction in one or more of the steps, including, without limitation, the intermediate pH maintenance step and the assembly step in the presence or absence of heating. One can, for example, add a stabilizer to the hinge-containing polypeptide to prevent or reduce aggregation. In other examples, a stabilizer can be added to the assembly mixture to prevent or reduce aggregation during the assembly of a heteromultimeric protein. In certain specific embodiments, the hinge-containing polypeptide comprises a half-antibody.

[047] In certain specific embodiments, the stabilizer is selected from the group consisting of arginine, histidine and polyvinylpyrrolidone (PVP). In certain embodiments, the arginine or histidine is an arginine salt or a histidine salt. In some other embodiments, arginine or histidine is a derivative of arginine or histidine. In some other embodiments, the arginine or histidine is arginine HCl or histidine HCl. In certain embodiments, arginine is not arginine phosphate.

[048] In some other embodiments, the method additionally comprises the step of incubating the assembly mixture in the presence of PVP. In related achievements, the PVP is added up to 40% (w/v). In some other embodiments, PVP is present in the assembly mixture at a concentration of 2% to 6% (w/v), 10% to 20%, 2% to 10%, 1%, 1.3%, 1, 7%, 2%, 2.3%, 2.7%, 3%, 3.3%, 3.7% or 4%, preferably from 0.1% to 10%, most preferably from 2% to 6 % and preferably higher than 4%. In certain embodiments, the PVP is no more than 100 kD, no more than 30 kD, and preferably 10 kD. In some other embodiments, PVP is present at less than 10% (w/v) or less than 5% (w/v).

[049] In some embodiments, the stabilizer is arginine present at a concentration of 20 mM to 1 mM, 20 mM to less than 1 mM, 20 mM to 100 mM, 20 mM to 200 mM, 20 mM to 300 mM, 20 mM to 400mM, 20mM to 50mM, 50mM to 100mM, 50mM to 150mM, 50mM to 200mM, 50mM to 250mM or 50mM to 300mM. In other embodiments, the stabilizer is histidine present at a concentration of 20 mM to 1 mM, 20 mM to less than 1 mM, 20 mM to 100 mM, 20 mM to 200 mM, 20 mM to 300 mM, 20 mM to 400 mM , 20mM to 50mM, 50mM to 100mM, 50mM to 150mM, 50mM to 200mM, 50mM to 250mM or 50mM to 300mM. In certain preferred embodiments, arginine or histidine is added at a concentration of 50 mM or 200 mM. In some other embodiments, arginine and/or histidine are added at a concentration of 20 mM to 200 mM, 20 mM to 100 mM, 50 mM to 200 mM, or 50 mM to 100 mM. In some other embodiments, the hinge-containing polypeptide comprises a half-antibody.

[050] In yet another aspect, the present invention provides a host cell that expresses a hinge-containing polypeptide. In certain embodiments, the hinge-containing polypeptide comprises a half-antibody.

[051] In a further aspect, the present invention provides a method of producing bispecific antibodies, comprising the steps of (a) culturing a first host cell designed to express a first half-antibody specific for a first antigen or first epitope of an antigen; (b) culturing a second host cell designed to express a second half-antibody specific for a second antigen or second epitope of the same antigen; (c) obtaining the first antibody medium from the cultivation of step a under pH 4 to 9, preferably 5 to 9, in the presence of a first solubilizer; (d) obtaining the second antibody medium from the cultivation of step b under pH 4 to 9, preferably 5 to 9, in the presence of a second solubilizer; (e) mixing the first and second half-antibodies in a reducing condition to form an assembly mixture; and (f) incubating the assembly mixture to form a bispecific antibody that comprises the first and second half-antibodies.

[052] In some embodiments, the first host cell and the second host cell are cultured in separate cultures, and the first half-antibody and the second half-antibody are purified separately from the cultures of the first and second host cells prior to mixing. In certain embodiments, the first and second host cells are grown in separate cultures, the cultures are combined, the cells are pelleted, optionally homogenized and/or lysed, and the first and second antibody media are co-purified by any appropriate methods. In certain embodiments, the first and second half-antibodies are co-purified by means of protein A purification. In further embodiments, the first and second host cells are co-cultured in a mixed culture and the first and second half-antibodies are co-purified.

[053] The host cell can, for example, be a bacterial cell, a yeast cell, a plant cell, an insect cell or a mammalian cell. In some specific embodiments, the hinge- or half-antibody containing polypeptide is produced by a mammalian cell, such as a CHO cell. In some other embodiments, the host cell is a bacterial cell, particularly E. coli.

[054] In certain further embodiments, the methods according to the present invention further comprise the step of recovering the heteromultimeric protein or bispecific antibody formed in step d. The assembled heteromultimeric protein can be further purified by methods described throughout the application or appropriate methods known in the art.

[055] In a further aspect, the present invention provides compositions comprising a polypeptide containing joint and a solubilizer, wherein the pH of the composition is from pH 4 to pH 9, preferably pH 5-9. In certain embodiments, the pH of the composition is at least pH 5, at least pH 5.5, at least pH 5.7, pH greater than 5, pH greater than 5.5 pH greater than 5.7, from 5 to 9, 5 to 8, 5.5 to 8, 5.5 to 9, 5.7 to 8, 5.7 to 9, 6 to 8, 6 to 9, 7 to 8, 7.5 to 8.5 or 7 to 8.5. In certain embodiments, the solubilizer is selected from the group consisting of arginine, histidine and sucrose, preferably arginine and/or histidine. In some other embodiments, arginine is a salt of arginine and/or histidine is a salt of histidine. In some other embodiments, arginine is a derivative of arginine and/or histidine is a derivative of histidine. In some other embodiments, the arginine or histidine is L-arginine or L-histidine. In some other embodiments, the arginine or histidine is arginine HCl or histidine HCl.

[056] In some specific embodiments, the solubilizer is arginine or histidine. In some other embodiments, arginine or histidine is present at a concentration of 20mM, 50mM, 200mM, 400mM, 20mM to 1M, 20mM less than 1M, 20mM to 200mM, 20mM to 400mM, 20mM to 100mM, 50mM to 100mM, 50mM to 200mM, 50mM to 300mM or 50mM to 400mM. In some specific embodiments, the composition comprises arginine and histidine, each present at a concentration of 20mM, 50mM, 200mM, 400mM, 20mM to 1M, 20mM less than 1M, 20mM to 20mM , 20mM to 400mM, 20mM to 100mM, 50mM to 100mM, 50mM to 200mM, 50mM to 300mM or 50mM to 400mM. In some other embodiments, the composition does not comprise guanidine HCl or urea. In certain embodiments, the composition alternatively or additionally comprises a stabilizer.

[057] In certain specific embodiments, the hinge-containing polypeptide comprises a half-antibody. In some other specific embodiments, the composition comprises only one type of hinge-containing polypeptide or half-antibody. In certain embodiments, the composition comprises only one type of half-antibody which is a half-bulge antibody (knob). In some other embodiments, the composition comprises only one type of half antibody which is a hole half antibody.

[058] In some specific embodiments, the composition further comprises a second hinge-containing polypeptide, wherein the first hinge-containing polypeptide comprises a bulge (knob) and the second hinge-containing polypeptide comprises a hole (hole). In certain embodiments, the hinge-containing polypeptide comprises a half-antibody. In certain specific embodiments, the hinge-containing polypeptide is a half-antibody. In some other embodiments, the IgG half-antibody is of the IgG1, IgG2 or IgG4 isotype.

[059] All realizations described herein may be combined, unless the context clearly indicates otherwise. Furthermore, any and all embodiments described above apply to any and all aspects of the present invention, unless the context clearly indicates otherwise.

[060] Other objects, features and advantages of the present invention will become evident from the detailed description below. It should be understood, however, that the detailed description and specific examples, while indicating preferred embodiments of the present invention, are provided by way of illustration only, as various changes and modifications within the scope and spirit of the present invention become will become apparent to those skilled in the art from the present detailed description.

BRIEF DESCRIPTION OF THE FIGURES

[061] Figure 1A illustrates a fully oxidized half-antibody. The “knob”, “hole” or other domains of heterodimerization are not shown. The half-antibody illustrated in this figure is an IgG1 isotype. Those skilled in the art will appreciate that the other immunoglobulin isotypes can be thought of as half-antibodies with the corresponding inter- and intra-strand linkages. In an intact antibody, the hinge cysteines will form interchain disulfide bonds.

[062] Figure 1B illustrates a full-length bispecific antibody with a heteromultimerization domain. Inter-heavy chain disulfide bonds in the hinge region are not shown. The heteromultimerization domain shown is the knob-in-hole shape.

[063] Figure 1C is a drawing representation of a bispecific antibody comprising a heteromultimerization domain (knob-in-hole), a furin-divisible teter and an optional additional disulfide bond (S354) . Also shown are the inter-heavy chain disulfide bonds in the hinge region. Furin split locations are indicated by triangles. Although the furin-divisible teter is displayed over the half-antibody that comprises the knob, it can also be used over the hole half-antibody or over the two half-antibodies, knob and of hole (hole).

[064] Figures 2A and B show composite size exclusion chromatograms demonstrating the effects of pH on the conformational change of half-antibodies. Figure 2A demonstrates that high pH induced the conformational change of the half-hole antibody that results in a larger hydrodynamic radius. This conformational change increased heterodimerization during assembly. Figure 2B demonstrates that high pH promoted the formation of non-covalent bump antibody-half homodimer. This change in conformation favored the bispecific formation during assembly. Reference is made to Example 2.

[065] Figures 3A and 3B illustrate the results demonstrating that a solubilizer such as arginine (Figures 3A and B) and histidine hydrochloride (Fig. 3B) reduced the intermediate pH-induced precipitation of medium-bulge antibodies (knob) .

[066] Figures 4A and B show composite chromatograms demonstrating the effect of a reducer such as glutathione on the aggregation and assembly of bispecific antibodies. Glutathione is added in 2 to 200X molar excess. Reference is made to Example 3.

[067] Figure 5A is a graph illustrating the effect of temperature on the rate of formation of bispecific IgG1 antibodies (assembly). Figure 5B demonstrates that higher temperature promoted the assembly of knob-in-hole bispecific IgG1 antibody in the presence of 200x molar excess of glutathione with or without pH maintenance, as analyzed by phase chromatography reverse. Figure 5B also illustrates that optimizing the pH maintenance of the half-antibody assemblies to address conformational changes prior to assembly increased assembly speed and efficiency. Reference is made to Example 4.

[068] Figure 6A illustrates the effect of a stabilizer such as PVP on the stabilization of the formed bispecific antibody and the reduction of the formation of aggregates.

[069] Figure 6B demonstrates that PVP and histidine under elevated temperature promoted the assembly of knob-in-hole bispecific IgG4 antibody as analyzed by reversed-phase chromatography. Lower curve: room temperature setup with 300x Glutathione:Ab ratio, pH = 8.5, about 400 mM arginine; upper curve: heated mount with 200x Glutathione:Ab ratio, pH = 8.0, 4% PVP, 50 mM arginine, 100 mM histidine at 35 °C.

[070] Figure 6C demonstrates that PVP reduced IgG4 knob-in-hole antibody aggregation during heated mounting at 37 °C and pH 8 for six hours. Reference is made to Example 5.

[071] Figure 7 illustrates that histidine reduced heat-induced aggregation of IgG4 half-hole antibodies at 37 °C and pH 8 for three hours.

DETAILED DESCRIPTION OF THE INVENTION

[072] The present invention will now be described in detail for reference only, using the definitions and examples below. All patents and publications, including all sequences disclosed in those patents and publications referenced herein are expressly incorporated by reference.

[073] Unless otherwise defined herein, all technical and scientific terms used herein have the same meaning commonly understood by those of ordinary skill in the art to which this invention belongs. Singleton et al, Dictionary of Microbiology and Molecular Biology, Second Edition, John Wiley and Sons, New York (1994) and Hale & Marham, The Harper Collins Dictionary of Biology, Harper Perennial, New York (1991) provide practitioners in the field with a general dictionary of many of the terms used in the present invention. While any methods and materials similar or equivalent to those described herein may be used in the practice or testing in accordance with the present invention, preferred methods and materials are described. Numeric ranges include the numbers that define the range. Unless otherwise noted, nucleic acids are written left to right in 5' to 3' orientation; amino acid sequences are written left to right in amino to carboxyl orientation, respectively. Practitioners are particularly advised to consult Sambrook et al, 1989, and Ausubel, F.M. et al, 1993, for definitions and terms of the technique. It should be understood that the present invention is not limited to the specific methodology, protocols and reagents described as these may vary.

[074] Numeric ranges include the numbers that define the range.

[075] Unless otherwise noted, nucleic acids are written left to right in 5' to 3' orientation; and the amino acid sequences are written from left to right in amino to carboxyl orientation, respectively.

[076] As used herein, the singular forms "a", "an" and "the" include plural references, unless the context clearly indicates otherwise. Reference to "a host cell", for example, indicates one or more host cells.

[077] The headings provided herein are not limitations of the various aspects or embodiments of the present invention which may be obtained by reference to the descriptive report as a whole. Consequently, the terms defined immediately below are more fully defined by referring to the descriptive report as a whole.

[078] All realizations described herein may be combined, unless the context clearly indicates otherwise. Furthermore, any and all embodiments described above apply to any and all aspects of the present invention, unless the context clearly indicates otherwise.

[079] I. Definitions: The present invention provides methods of producing heteromultimeric proteins comprising a first hinge-containing polypeptide and a second hinge-containing polypeptide. The term "hinge-containing polypeptide" as used herein means a polypeptide which contains at least one hinge region. In certain embodiments, the hinge region connects several domains, such as a binding domain and an effector domain, and provides some structural flexibility to the polypeptide for dimerization or multimerization. As an example, the binding domain can be an antigen binding domain of an antibody or a ligand binding domain of a receptor and the effector domain can be an Fc component of an antibody. In certain embodiments, the first hinge-containing polypeptide is different from the second hinge-containing polypeptide and the resulting dimer or multimer is a heterodimer or heteromultimer. In certain specific embodiments, the first hinge-containing polypeptide and the second hinge-containing polypeptide bind to two different epitopes on the same target protein. In some other embodiments, the first hinge-containing polypeptide has different target binding specificity than the second hinge-containing polypeptide and the resulting heterodimer or heteromultimer binds to two or more different target proteins. In certain embodiments, the linkage-containing polypeptide comprises a naturally occurring or elaborate heterodimerization domain. In certain specific embodiments, the hinge-containing polypeptide comprises one or more cysteine residues in the hinge region capable of forming one or more disulfide bonds with another hinge-containing polypeptide.

[080] A hinge-containing polypeptide includes, without limitation, a half-antibody, an immunoadhesin and its functional fragment. The term "functional fragment", as used herein, means a fragment, that is, less than the full length of the hinge-containing polypeptide, which still maintains the desired function; it retains, for example, antigen-binding or targeting activity, Fc effector activity, and/or dimerization/multimerization capacity. In certain specific embodiments, each of the first hinge-containing polypeptide and the second hinge-containing polypeptide is a half-antibody with different antigen-binding specificity, and the resulting dimer or multimer is a bispecific or multispecific antibody. In certain embodiments, the resulting heteromultimeric protein comprises a half-antibody and an immunoadhesin.

[081] The term "multispecific antibody" is used in the broadest sense and specifically covers an antibody that has polyepitopic specificity. Such multispecific antibodies include, but are not limited to, an antibody comprising a heavy chain variable domain (VH) and a light chain variable domain (VL), wherein the VHVL unit has polyepitopic specificity, antibodies that have two or more VL domains and VH where each VHVL unit binds to a different epitope, antibodies which have two or more isolated variable domains where each isolated variable domain binds to a different epitope, full length antibodies, antibody fragments such as Fab, Fv, dsFv, scFv, specific diabodies, diabodies and triabodies, and antibody fragments that have been linked covalently or non-covalently. "Polyepitopic specificity" means the ability to specifically bind to two or more different epitopes on the same or different target(s). "Monospecific" means the ability to bind only one epitope. In one embodiment, the multispecific antibody is an IgG antibody that binds to each epitope with an affinity of 5 μM to 0.001 pM, 3 μM to 0.001 pM, 1 μM to 0.001 pM, 0.5 μM to 0.001 pM or 0.1 μM at 0.001 pM.

[082] A naturally occurring four-chain antibody basic unit is a heterotetrameric glycoprotein composed of two identical light (L) chains and two identical heavy (H) chains (an IgM antibody consists of five of the basic heterotetrameric units, along with one additional polypeptide called J-chain, therefore containing ten antigen-binding sites, whereas secreted IgA antibodies can polymerize to form polyvalent assemblies comprising two to five of the basic four-chain units along with the J-chain). In the case of IgGs, the four-chain unit is usually about 150,000 daltons. Each L chain is linked to an H chain by a covalent disulfide bond, while the two H chains are linked together by one or more disulfide bonds, depending on the isotype of the H chain. Each H and L chain also contains disulfide bridges intrachains in regular spaces. Each H chain contains, at the N-terminus, a variable domain (VH) followed by three constant domains (CH) for each of the α and Y chains and four CH domains for the μ and ε isotypes. Each L chain contains, at the N-terminus, a variable domain (VL) followed by a constant domain (CL) at the C-terminus. The VL is aligned to the VH and the CL is aligned to the first constant domain of the heavy chain (CH1). It is believed that specific amino acid residues form an intermediate surface between the light chain and heavy chain variable domains. The pairing of VH and VL with each other forms an isolated antigen-binding site. For structure and properties of the different classes of antibodies, see, for example, Basic and Clinical Immunology, 8th edition, Daniel P. Stites, Abba I., Terr and Tristam, G. Parslow (Eds.), Appleton & Lange, Norwalk CT , 1994, p. 71 and Chapter 6.

[083] The L chain of any vertebrate species can be classified into one of two clearly distinct types, termed kappa and lambda, based on the amino acid sequences of its constant domains. Depending on the amino acid sequence of the constant domain of their heavy chains (CH), immunoglobulins can be assigned to different classes or isotypes. There are five classes of immunoglobulins: IgA, IgD, IgE, IgG and IgM, which contain heavy chains called α, δ, ε, Y and μ, respectively. Classes Y and α are further divided into subclasses based on relatively minor differences in CH function and sequence; humans, for example, express the following subclasses: IgG1, IgG2, IgG3, IgG4, IgA1 and IgA2.

[084] The term "variable" designates the fact that certain segments of the variable domains differ extensively in sequence among antibodies. Domain V mediates antigen binding and defines the specificity of a specific antibody for its specific antigen. The variability is not evenly distributed, however, across the 110-amino acid series of the variable domains. Instead, V regions consist of relatively invariant stretches called framework regions (FRs) of fifteen to thirty amino acids separated by shorter regions of extreme variability called “hypervariable regions,” which contain nine to twelve amino acids in length each. Each of the heavy and light chain variable domains comprises four FRs, which largely adopt a beta-sheet configuration, connected by three hypervariable regions, which form circuits that connect and, in some cases, form part of the beta-sheet structure. The hypervariable regions of each chain are held together in close proximity by the FRs and, with the hypervariable regions of the other chain, contribute to antibody antigen binding site formation (see Kabat et al, Sequences of Proteins of Immunological Interest, fifth edition, Public Health Service, National Institutes of Health, Bethesda MD (1991)). Constant domains are not directly involved in the binding of an antibody to an antigen, but they exhibit several effector functions, such as antibody participation in antibody-dependent cellular cytotoxicity (ADCC).

[085] The expression "hypervariable region", "HVR" or "HV", as used herein, indicates regions of an antibody variable domain that are of hypervariable sequence and/or form structurally defined circuits. Antibodies generally comprise six HVRs; three in VH (H1, H2, H3) and three in VL (L1, L2, L3). In native antibodies, H3 and L3 exhibit the greatest diversity of the six HVRs and H3 is believed to play a particularly unique role in providing fine specificity to antibodies. See, for example, Xu et al, Immunity 13: 37-45 (2000); Johnson and Wu, in Methods in Molecular Biology 248: 1-25 (Lo, ed., Human Press, Totowa NJ, 2003). In fact, naturally occurring camelid antibodies that consist of a single heavy chain are functional and stable in the absence of a light chain. See, for example, Hamers-Casterman et al, Nature 363: 446448 (1993); Sheriff et al, Nature Struct. Biol. 3: 733-736 (1996).

[086] Several delineations of HVR are in use and are encompassed at present. Kabat Complementarity Determining Regions (CDRs) are based on sequence variability and are the most commonly used (Kabat et al, Sequences of Proteins of Immunological Interest, fifth edition, Public Health Service, National Institutes of Health, Bethesda MD (1991)). Chothia refers, in turn, to the location of structural circuits (Chothia and Lesk, J. Mol. Biol. 196: 901-917 (1987)). AbM HVRs represent a compromise between Kabat HVRs and Chothia structural circuitry and are used by Oxford Molecular's AbM antibody modeling software. “Contact” HVRs are based on analysis of the complex crystal structures available. Residues from each of these HVRs are indicated below.

[087] HVRs may comprise "Extended HVRs" as follows: 24-36 or 24-34 (L1), 46-56 or 50-56 (L2) and 89-97 or 89-96 (L3) in the VL and 26 -35 (H1), 50-65 or 49-65 (H2) and 93-102, 94-102 or 95-102 (H3) in the VH. Variable domain residues are numbered according to Kabat et al, above, for each of these definitions.

[088] "Framework regions" (FR) are variable domain residues other than CDR residues. Each variable domain typically contains four FRs identified as FR1, FR2, FR3 and FR4. If the CDRs are defined according to Kabat, the light chain FR residues are positioned near residues 1 to 23 (LCFR1), 35 to 49 (LCFR2), 57 to 88 (LCFR3) and 98 to 107 (LCFR4) and heavy chain FR residues are positioned near residues 1 to 30 (HCFR1), 36 to 49 (HCFR2), 66 to 94 (HCFR3) and 103 to 113 (HCFR4) in the heavy chain residues. If the CDRs comprise hypervariable circuitry amino acid residues, the light chain FR residues are positioned near residues 1 to 25 (LCFR1), 33 to 49 (LCFR2), 53 to 90 (LCFR3), and 97 to 107 (LCFR4) in the light chain and the heavy chain FR residues are positioned near residues 1 to 25 (HCFR1), 33 to 52 (HCFR2), 56 to 95 (HCFR3) and 102 to 113 (HCFR4) in the heavy chain residues. In some cases, when the CDR comprises amino acids from a CDR as defined by Kabat and those from a hypervariable circuit, the FR residues will be adjusted accordingly. When CDRH1 includes amino acids H26 to H35, for example, heavy chain FR1 residues are found in positions 1 to 25 and FR2 residues are found in positions 36 to 49.

[089] "Human consensus framework" is a framework representing the most commonly occurring amino acid residues in a selection of human immunoglobulin VL or VH framework sequences. Generally, the selection of human immunoglobulin VL or VH sequences is from a subset of variable domain sequences. Generally, the sequence subgroup is a subgroup as in Kabat et al. In one embodiment, for the VL, the subgroup is cape I subgroup as in Kabat. In one embodiment, for VH, the subgroup is subgroup III as in Kabat.

[090] An example of an "intact" antibody is one that comprises an antigen binding site as well as a CL and at least constant heavy chain domains, CH1, CH2 and CH3. The constant domains can be native sequence constant domains (for example, human native sequence constant domains) or amino acid sequence variants thereof.

[091] "Antibody fragments" comprise a portion of an intact antibody, preferably the variable or antigen binding region of the intact antibody. Examples of antibody fragments include Fab, Fab', F(ab')2 and Fv fragments; diabodies (Db); diabodies together (taDb), linear antibodies (e.g., U.S. Patent No. 5,641,870, Example 2; Zapata et al, Protein Eng. 8(10): 1057-1062 (1995)); single-arm antibodies, single variable domain antibodies, minibodies, single-chain antibody molecules; and multispecific antibodies formed with antibody fragments (including, for example, but not limited to, Db-Fc, taDb-Fc, taDb-CH3 and (scFV)4-Fc).

[092] The term "half-antibody", as used herein, designates an immunoglobulin heavy chain associated with an immunoglobulin light chain. An example half-antibody is illustrated in Figure 1A. Those skilled in the art will readily appreciate that a half-antibody may encompass one of its fragments and may also have an antigen-binding domain consisting of a single variable domain originating, for example, from a camelid.

[093] The present inventors unexpectedly discovered that the optimization or pH adjustment of half-antibodies eluted from a column of Protein A or other matrix under low pH induced conformational alteration of hinge-containing polypeptides such as half-antibodies. Optimizing pH at mid-pH, sometimes referred to as pH maintenance or mid-pH maintenance throughout this report, can cause precipitation or aggregation of the media-antibodies. Thus, in certain embodiments, the method of producing a heteromultimeric protein comprises the step of providing a first or second hinge-containing polypeptide under pH 5-9 in the presence of a first or second solubilizer, respectively.

[094] A solubilizer, as used herein, is defined as a reagent that prevents or reduces the precipitation of a hinge-containing polypeptide, such as a half-antibody. A suitable solubilizer includes, without limitation, arginine and histidine or one of their salts or derivatives and sucrose. In some embodiments, the solubilizer is arginine and/or histidine. In certain embodiments, the solubilizer prevents or reduces precipitation induced by maintaining an intermediate pH and/or heating. In certain specific embodiments, a solubilizer is added prior to maintaining the intermediate pH (i.e., prior to adjusting to the intermediate pH) and/or heating. In certain embodiments, the arginine or histidine is an arginine salt or a histidine salt. In some other embodiments, the arginine or histidine is an arginine derivative or histidine derivative. Reduced precipitation can increase the yield of the desired final assembled product.

[095] Imidazole and guanidine have been used to solubilize proteins for general protein preparation and purification. It was unexpectedly found, however, that imidazole and guanidine alone, outside the context of histidine or arginine, respectively, were insufficient to increase the overall yield of assembled heteromultimeric protein as described herein, such as a bispecific antibody. In certain embodiments, imidazole and guanosine can denature proteins.

[096] Similarly, detergents such as guanidine HCl and urea are commonly used to reduce overall aggregation/precipitation, but can completely denature the protein. Thus, in certain embodiments, the solubilizer prevents or reduces precipitation without denaturing the protein of interest. Thus, in certain specific embodiments, the solubilizer is not guanidine HCl, guanidine, imidazole or urea. In some other embodiments, the compositions according to the present invention do not comprise guanidine HCl or urea. In some other embodiments, the solubilizer is neither Tween nor PEG.

[097] In addition, a stabilizer can be added, for example, during maintenance of intermediate pH of each half-antibody or during assembly in the mixture of polypeptides that contain articulation or half-antibody or shortly thereafter. A stabilizer may be added to the reaction in one or more or all of the steps of the methods according to the present invention to prevent or reduce aggregation of the hinge-containing polypeptides or half-antibodies before, during and/or after assembly.

[098] Aggregates can be detected as substances with high molecular weight and, in the context of half-antibodies, substances with high molecular weight of more than 150 kDa. Aggregates can be detected and quantified, for example, by size exclusion chromatography or other appropriate methods. In some other embodiments, aggregates detected via size exclusion chromatography can pass through a sterile 0.2 µm filter. Precipitated proteins, on the other hand, can be composed of denatured proteins or aggregated proteins that can form very large complexes. Several reagents have been tested and found to be ineffective or unsuitable for use as stabilizers, such as imidazole, 3-(N-morpholino) propanesulfonic acid (MOPS), 2-(N-morpholino)ethanesulfonic acid (MES), cyclodextrin, CuSO4 and Not AC. Thus, in certain embodiments, the stabilizer does not include inorganic salts or transition metals. Suitable stabilizers include, without limitation, PVP, histidine and arginine. Reduction of aggregation can generate increased yield of the desired final assembled product.

[099] PVP is a water-soluble uncharged polymer with a pyrrolidone group. In certain embodiments, other uncharged polar polymers, other reagents or compounds, especially compounds with similar structure and properties to the appropriate stabilizers described herein may be suitable stabilizers for use in the present invention. It is within the skill of those skilled in the art to determine an appropriate stabilizer by analyzing the effect of the compound on aggregation levels by methods known in the art, including those provided herein.

[0100] A reagent can be characterized as solubilizer and stabilizer. Arginine, for example, can be used as a solubilizer to reduce precipitation of half-antibodies during maintenance of intermediate pH and/or heating, as well as a stabilizer to reduce aggregation during the assembly step. Similarly, histidine can be used as a solubilizer to reduce precipitation, as well as a stabilizer to reduce media-antibody aggregation during maintenance of intermediate pH and/or heating. Without wishing to be limited to any specific mechanism, in certain embodiments, a solubilizer and a stabilizer can function by preventing the interaction of hydrophobic patches on protein surfaces that can generate aggregation. In other embodiments, a solubilizer and a stabilizer can work to form clathrates to prevent unwanted protein interaction.

[0101] The term "single-chain half-antibody", as used herein, means a single-chain polypeptide comprising a VL domain, optionally a CL domain, a tether, a VH domain, optionally a CH1 domain, a hinge domain, a CH2 domain and a CH3 domain, wherein said domains are positioned together in the N-terminal to C-terminal direction as follows: VL-teter-VH-joint-CH2-CH3 or VL-CL-teter -VH-CH1-joint-CH2-CH3.

[0102] The term "isolated domain antibodies" (sdAbs) or "isolated variable domain antibodies (SVD)" generally designates antibodies in which a single variable domain (VH or VL) can confer antigen binding. In other words, the isolated variable domain does not need to interact with another variable domain in order to recognize the target antigen. Examples of isolated domain antibodies include those derived from camelids (llamas and camels) and cartilaginous fish (such as sand sharks) and those derived from recombinant mouse and human antibody methods (Nature (1989) 341: 544-546; Dev Comp. Immunol. (2006) 30: 43-56; Trend Biochem. Sci. (2001) 26: 230-235; Trends Biotechnol. (2003): 21: 484-490; WO 2005/035572; WO 03/035694 ; Febs Lett. (1994) 339: 285-290; WO 00/29004; WO 02/051870).

[0103] The term "linear antibodies" generally designates the antibodies described in Zapata et al, Protein Eng. 8(10): 1057-1062 (1995). Briefly, these antibodies comprise a pair of together Fd segments (VH-CH1-VH-CH1) which, together with the complementary light chain polypeptides, form a pair of antigen binding regions. Linear antibodies can be bispecific or monospecific.

[0104] The term “knob-in-hole” or “KnH” technology, as mentioned herein, designates the technology that directs the pairing of two polypeptides together in vitro or in vivo through introduction of a bulge (knob) in one polypeptide and a cavity (hole - hole) in the other polypeptide at an interface at which they interact. KnHs, for example, have been introduced into Fc:Fc binding interfaces, CL:CH1 interfaces or VH/VL interfaces of antibodies (e.g. US 2007/0178552, WO 96/027011, WO 98/050431 and Zhu et al (1997) ), Protein Science 6: 781-788). This is especially useful for directing pairing of two different heavy chains together during the manufacture of multispecific antibodies. Multispecific antibodies that have KnH in their Fc regions can further comprise, for example, isolated variable domains linked to each Fc region, or further comprise different heavy chain variable domains that pair with similar or different light chain variable domains. KnH technology can also be used to pair two different receptor extracellular domains together or any other polypeptide sequences that comprise different target recognition sequences (including, for example, affinity bodies, peptide bodies and other Fc fusions).

[0105] Papain digestion of antibodies produces two identical antigen-binding fragments, termed “Fab” fragments, and a residual “Fc” fragment, whose designation reflects the ability to rapidly crystallize. The Fab fragment consists of an entire L chain, along with the variable region domain of the H chain (VH) and the first constant domain of a heavy chain (CH1). Pepsin treatment of an antibody generates a single large F(ab')2 fragment, which roughly corresponds to two disulfide-linked Fab fragments that have divalent antigen-binding activity and are still capable of cross-linking antigens. Fab' fragments differ from Fab fragments in that they contain some additional residues at the carboxy terminus of the CH1 domain, including one or more cysteines from the antibody hinge region. Fab'-SH is the present designation for Fab' in which the cysteine residue(s) of the constant domains bear(s) a free thiol group. F(ab')2 antibody fragments were originally produced as pairs of Fab' fragments that contain hinge cysteines between them. Other chemical couplings of antibody fragments are also known.

[0106] The Fc fragment comprises the carboxy-terminal parts of the two H chains held together by disulfides. Antibody effector functions are determined by sequences in the Fc region; this region is also the part recognized by Fc receptors (FcR) found on certain cell types.

[0107] "Fv" consists of a dimer of one light chain and one heavy chain variable region domain in tight, non-covalent association. From the fold of these two domains emanate six hypervariable loops (three H and L chain loops each) that contribute the amino acid residues for antigen binding and confer antigen-binding specificity to the antibody. However, even an isolated variable domain (or half of an Fv that comprises only three specific CDRs for an antigen) has the ability to recognize and bind antigen, although often at lower affinity than the complete binding site.

[0108] "Single-chain Fv", also abbreviated as "sFv" or "scFv", are antibody fragments that comprise the VH and VL antibody domains connected to a single polypeptide chain. Preferably, the sFv polypeptide further comprises a polypeptide linker between the VH and VL domains, which allows the sFv to form the desired structure for antigen binding. For sFv analysis, see Pluckthun, The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore (eds.), Springer-Verlag, New York, p. 269-315 (1994); Malmborg et al, J. Immunol. Methods 183:7-13, 1995.

[0109] The term "diabodies" designates small antibody fragments prepared by constructing sFv fragments (see previous paragraph) with short linkers (about five to ten residues) between the VH and VL domains, in order to target if the V domains interchain but not intrachain pairing, which results in a bivalent fragment, that is, a fragment that contains two antigen binding sites. Bispecific diabodies are heterodimers of two "cross-over" sFv fragments, in which the VH and VL domains of the two antibodies are present on different polypeptide chains. Diabodies are more fully described, for example, in EP 404,097, WO 93/11161 and Hollinger et al, Proc. Natl. Academic Sci. U.S.A. 90:6444-6448 (1993).

[0110] The term "one-armed antibody" or "one-armed antibodies" means an antibody comprising (1) a variable domain linked by a peptide bond to a polypeptide comprising a CH2 domain, a CH3 domain or a domain CH2-CH3 and (2) a second CH2, CH3 or CH2-CH3 domain, wherein a variable domain is not joined by a peptide bond to a polypeptide comprising the second CH2, CH3 or CH2-CH3 domain. In one embodiment, the one-armed antibody comprises three polypeptides: (1) a first polypeptide that comprises a variable domain (such as VH), CH1, CH2 and CH3; (2) a second polypeptide comprising a variable domain (such as VL) and a CL domain; and (3) a third polypeptide comprising a CH2 and CH3 domain. In one embodiment, the third polypeptide does not comprise a variable domain. In another embodiment, the one-armed antibody has a partial hinge region that contains the two cysteine residues that form disulfide bonds that link the constant heavy chains. In one embodiment, the one-armed antibody variable domains form an antigen-binding region. In another embodiment, a one-armed antibody variable domain is an isolated variable domain, wherein each isolated variable domain is an antigen binding region.

[0111] Antibodies according to the present invention can be "chimeric" antibodies, in which a part of the light and/or heavy chain is identical or homologous to corresponding sequences in antibodies derived from a specific species or belonging to a class or subclass antibody-specific, while the remainder of the chain(s) is identical or homologous to corresponding sequences of antibodies derived from other species or belonging to another class or subclass of antibodies, as well as fragments of these antibodies, provided that they exhibit biological activity desired (see U.S. Patent 4,816,567; and Morrison et al, Proc. Natl. Acad. Sci. USA, 81: 6851-6855 (1984)). Chimeric antibodies of interest herein include primatized antibodies which comprise variable domain antigen binding sequences derived from non-human primates (e.g., Old World Monkey, Chimpanzee etc.) and human constant region sequences.

[0112] "Humanized" forms of non-human antibodies (eg, from rodents) are chimeric antibodies that contain minimal sequence derived from the non-human antibody. Most humanized antibodies are human immunoglobulins (receptor antibody), in which residues from a hypervariable region of the recipient are replaced by residues from a hypervariable region from a non-human species (donor antibody), such as a mouse, rat, rabbit, or non-human primate that exhibits the desired antibody capacity, affinity, and specificity. In some cases, framework region (FR) residues of the human immunoglobulin are replaced by corresponding non-human residues. Furthermore, humanized antibodies may comprise residues that are not found in the recipient antibody or in the donor antibody. These modifications are made to further refine antibody performance. Generally, the humanized antibody will comprise substantially all of at least one, typically two variable domains, wherein all or substantially all of the hypervariable loops correspond to those of a non-human immunoglobulin and all or substantially all of the FRs are those of a human immunoglobulin sequence. The humanized antibody will optionally also comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. For further details, see Jones et al, Nature 321: 522-525 (1986); Riechmann et al, Nature 332: 323-329 (1988); and Presta, Curr. Op. Structure Biol. 2:593-596 (1992).

[0113] "Complex" or "in complex", as used herein, means the association of two or more molecules that interact with each other through bonds and/or forces (such as van der Waals, hydrophobic and hydrophilic forces ) which are not peptide bonds. In one embodiment, the complex is heteromultimeric. It should be understood that the term "protein complex" or "polypeptide complex" as used herein includes complexes that contain a non-protein entity conjugated to a protein in the protein complex (including, for example, but without limitation, chemical molecules such as a toxin or detection agent).

[0114] The term "heteromultimer" or "heteromultimeric", as used herein, describes two or more polypeptides that interact with each other by a non-peptide covalent bond (such as a disulfide bond) and/or a non-covalent interaction (such as hydrogen bonds, ionic bonds, Van der Waals forces and hydrophobic interactions), in which at least two of the molecules have different sequences from each other.

[0115] As used herein, "heteromultimerization domain" designates alterations or additions to a biological molecule, in order to promote the formation of heteromultimers and obstruct the formation of homomultimers. Any domain of heterodimerization that has a strong preference for forming heterodimers over homodimers is within the scope of the present invention. Illustrative examples include, but are not limited to, for example, U.S. Patent Application 20030078385 (Arathoon et al - Genentech; describes knob-in-holes); WO 2007147901 (Kjaergaard et al - Novo Nordisk; describes ionic interactions); WO 2009089004 (Kannan et al - Amgen; describes electrostatic targeting effects); WO 2010/034605 (Christensen et al - Genentech; describes wound spools). See also, for example, Pack, P. and Plueckthun, A., Biochemistry 31, 1579-1584 (1992), which describes leucine zipper, or Pack et al, Bio/Technology 11, 1271-1277 (1993), which describes the propeller-turn-helix motif. The expressions "heteromultimerization domain" and "heterodimerization domain" are used interchangeably at present. In certain embodiments, the linkage-containing polypeptide comprises one or more heterodimerization domains.

[0116] As used herein, the term "immunoadhesin" designates molecules that combine the binding specificity of a heterologous protein ("adhesin") with the effector functions of immunoglobulin constant domains. Structurally, immunoadhesins comprise an amino acid sequence fusion with the desired binding specificity, where the amino acid sequence is different from the antigen binding and recognition site of an antibody (i.e., is "heterologous" compared to a region antibody constant) and an immunoglobulin constant domain sequence (such as an IgG CH2 and/or CH3 sequence). Examples of adhesin sequences include contiguous amino acid sequences that comprise a receptor or linker portion that binds to a protein of interest. Adhesin sequences can also be sequences that bind to a protein of interest, but are not ligand or receptor sequences (such as adhesin sequences on peptide bodies). These polypeptide sequences can be selected or identified by a variety of methods, including phage display methods and high-throughput sorting methods. The immunoglobulin constant domain sequence in the immunoadhesin can be obtained from any immunoglobulin, such as the IgG-1, IgG-2, IgG-3 or IgG-4, IgA (including IgA-1 and IgA-2) subtypes. IgE, IgD or IgM.

[0117] An antibody according to the present invention "that binds" an antigen of interest is one that binds the antigen with sufficient affinity, such that the antibody is useful as a therapeutic and/or diagnostic agent to target a protein, cell, or tissue that expresses the antigen and does not significantly cross-react with other proteins. In these embodiments, the extent of antibody binding to a "non-target" protein will be less than about 10% of the antibody binding to its specific target protein, as determined by fluorescence activated cell ordering (FACS) analysis. , radioimmunoprecipitation (RIA) or ELISA. With respect to the binding of an antibody to a target molecule, the term "specific binding", "specifically binds to" or "is specific to" a specific polypeptide or epitope on a specific target polypeptide indicates binding that is measurably different from a non-specific interaction (a non-specific interaction, for example, can bind to casein or bovine serum albumin). Specific binding can be measured, for example, by determining the binding of a molecule compared to the binding of a control molecule. Specific binding can be determined, for example, by competition with a control molecule that is similar to the target, such as excess unlabeled target. In this case, specific binding is indicated if the binding of the labeled target to a probe is competitively inhibited by excess unlabeled target. The term "specific binding", "specifically binds to" or "is specific to" a specific polypeptide or an epitope on a specific target polypeptide, as used herein, may be exhibited, for example, by a molecule having Kd to target at least about 200 nM, alternatively at least about 150 nM, alternatively at least about 100 nM, alternatively at least about 60 nM, alternatively at least about 50 nM, alternatively at least about 40 nM, alternatively at least about 30 nM, alternatively at least about 20 nM, alternatively at least about 10 nM, alternatively at least about 8 nM, alternatively at least about 6 nM, alternatively at least about 4 nM, alternatively at least about 2 nM, alternatively at least about 1 nM or more. In one embodiment, the term "specific binding" indicates binding in which a molecule binds to a specific polypeptide or epitope on a specific polypeptide without substantial binding to any other polypeptide or polypeptide epitope.

[0118] "Binding affinity" generally designates the strength of the sum total of non-covalent interactions between a single binding site of a molecule (such as an antibody) and its binding partner (such as an antigen). Unless otherwise indicated, as used herein, "binding affinity" means intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (such as antibody and antigen). The affinity of an X molecule for its Y partner can generally be represented by the dissociation constant (Kd). Kd can be, for example, about 200nM, 150nM, 100nM, 60nM, 50nM, 40nM, 30nM, 20nM, 10nM, 8nM, 6nM, 4nM, 2nM, 1nM or more. Affinity can be measured by common methods known in the art, including those described herein. Low affinity antibodies generally bind antigen slowly and tend to dissociate quickly, whereas high affinity antibodies generally bind antigen more quickly and tend to stay bound longer. A number of methods of measuring binding affinity are known in the art and any of them can be used for the purposes of the present invention.

[0119] In one embodiment, "Kd" or "Kd value" in accordance with the present invention is measured using surface plasma resonance tests employing BIAcore® 2000 or BIAcore® 3000 (BIAcore, Inc., Piscataway NJ) at 25°C °C with CM5 antigen chips immobilized at about ten response units (RU). Briefly, carboxymethylated dextran biosensor chips (CM5, BIAcore Inc.) are activated with N-ethyl-N'-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) according to supplier instructions. Antigen is diluted with 10 mM sodium acetate, pH 4.8, in 5 μg/ml (about 0.2 μM) prior to injection at a flow rate of 5 μl/minute to achieve approximately ten response units ( RU) of coupled protein. After antigen injection, 1 M ethanolamine is injected to block unreacted groups. For kinetic measurements, serial two-fold dilutions of Fab (eg, 0.78 nM to 500 nM) are injected into PBS with 0.05% Tween 20 (PBST) at 25 °C under a flow rate of about 25 µl /min. Association rates (kon) and dissociation rates (koff) are calculated using a simple one-to-one Langmuir binding model (BIAcore Assessment Software version 3.2) through simultaneous configuration of the association and dissociation sensorgram. The equilibrium dissociation constant (Kd) is calculated as the ratio koff/kon. See, for example, Chen et al, J. Mol. Biol. 293: 865-881 (1999). If the on-rate exceeds 106 M-1 s-1 using the surface plasma resonance test above, the on-rate can then be determined using a fluorescent cooling method that measures the increase or decrease in the intensity of Fluorescence emission (excitation = 295 nm; emission = 340 nm, 16 nm bandpass) at 25 °C of 20 nM anti-antigen antibody (Fab form) in PBS, pH 7.2, in the presence of increasing concentrations of antigen as measured in a spectrometer, such as a flow-stop equipped spectrometer (Aviv Instruments) or SLM-AMINCO® 8000 series spectrophotometer (ThermoSpectronic) with a red shaken cuvette.

[0120] "Bond rate", "association rate" or "kon" according to the present invention can also be determined with the same surface plasma resonance method described above using BIAcore® 2000 or BIAcore® 3000 (BIAcore, Inc., Piscataway NJ) at 25°C with CM5 antigen chips immobilized at about ten response units (RU). Briefly, carboxymethylated dextran biosensor chips (CM5, BIAcore Inc.) are activated with N-ethyl-N'-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) according to supplier instructions. Antigen is diluted with 10 mM sodium acetate, pH 4.8, to 5 μg/ml (approximately 0.2 μM) before injection at a flow rate of 5 μl/minute to achieve approximately ten response units ( RU) of coupled protein. After antigen injection, 1 M ethanolamine is injected to block unreacted groups. For kinetic measurements, serial two-fold dilutions of Fab (eg, 0.78 nM to 500 nM) are injected into PBS with 0.05% Tween 20 (PBST) at 25 °C under a flow rate of about 25 µl /min. Association rates (kon) and dissociation rates (koff) are calculated using a simple one-to-one Langmuir binding model (BIAcore Assessment Software version 3.2) through simultaneous configuration of the association and dissociation sensorgram. The equilibrium dissociation constant (Kd) is calculated as the ratio koff/kon. See, for example, Chen et al, J. Mol. Biol. 293: 865-881 (1999). If the on-rate exceeds 106 M-1 s-1 using the surface plasma resonance test above, the on-rate can then be determined using a fluorescent cooling method that measures the increase or decrease in the intensity of Fluorescence emission (excitation = 295 nm; emission = 340 nm, 16 nm bandpass) at 25 °C of 20 nM anti-antigen antibody (Fab form) in PBS, pH 7.2, in the presence of increasing concentrations of antigen as measured in a spectrometer, such as a spectrometer equipped with a stop flow (Aviv Instruments) or an SLM-Aminco® 8000 series spectrophotometer (ThermoSpectronic) with an agitated cuvette.

[0121] "Biologically active", "biological activity" and "biological characteristics", with respect to a polypeptide according to the present invention, such as its antibody, fragment or derivative, indicates that it has the ability to bind to a biological molecule, unless otherwise specified.

[0122] "Peptide body" or "peptide bodies" designates a fusion of randomly generated peptides with an Fc domain. See U.S. Patent No. 6,660,843, issued December 9, 2003 to Feige et al (incorporated in its entirety by reference). They include one or more peptides linked to the N-terminus, C-terminus, amino acid side chains, or more than one of these sites. Peptide body technology allows for the design of therapeutic agents that incorporate peptides that target one or more ligands or receptors, tumor harboring peptides, membrane transporting peptides, and the like. Peptide body technology has been shown to be useful in the design of a number of these molecules, including linear and disulfide-restricted peptides, "collective peptide multimers" (ie, more than one peptide over a single strand of a domain Fc). See, for example, U.S. Patent No. 6,660,843; U.S. Patent Application No. 2003/0195156, published October 16, 2003 (corresponding to WO 02/092620, published November 21, 2002); U.S. Patent Application No. 2003/0176352, published September 18, 2003 (corresponding to WO 03/031589, published April 17, 2003); U.S. Serial Number 09/422,838, filed October 22, 1999 (corresponding to WO 00/24770, published May 4, 2000); U.S. Patent Application No. 2003/0229023, published December 11, 2003; WO 03/057134, published July 17, 2003; U.S. Patent Application No. 2003/0236193, published December 25, 2003 (corresponding to PCT/US04/010989, filed April 8, 2004); US Serial Number 10/666,480, filed September 18, 2003 (corresponding to WO 04/026329, published April 1, 2004), each of which is incorporated herein by reference in its entirety.

[0123] "Affinity body" or "affinity bodies" designates the use of a protein linked by peptide binding to an Fc region, where the protein is used as a scaffold to provide a binding surface for a target molecule. The protein is often a naturally occurring protein such as staphylococcus protein A or IgG binding domain B, or the protein Z derived therefrom (see Nilsson et al (1987), Prot. Eng. 1, 107-133 and the Patent US No. 5,143,844) or one of its fragments or derivatives. Affinity bodies can be created, for example, from Z protein variants that have altered binding affinity for target molecule(s), where a segment of the Z protein has been mutated through random mutagenesis to create a library of variants capable of binding a target molecule. Examples of affinity bodies include U.S. Patent No. 6,534,628, Nord, K. et al, Prot. Eng. 8: 601-608 (1995) and Nord, K. et al, Nat. Biotech. 15: 772-777 (1997), Biotechnol. Appl. Biochem., June 2008; 50 (Part 2): 97-112.

[0124] "Isolated" or complex heteromultimer indicates a heteromultimer or complex that has been separated and/or recovered from a component of its natural cell culture environment. Contaminant components of its natural environment are materials that would interfere with diagnostic or therapeutic uses for the heteromultimer and may include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. In preferred embodiments, the heteromultimer will be purified (1) to greater than 95% by weight protein, as determined by the method of Lowry, and more preferably greater than 99% by weight; (2) to a degree sufficient to obtain at least fifteen residues of internal or N-terminal amino acid sequence using a centrifuge cup sequencer; or (3) to homogeneity, by means of SDS-PAGE, under reducing or non-reducing conditions, using Coomassie blue or, preferably, silver stains.

[0125] Heteromultimers according to the present invention are generally purified to substantial homogeneity. The terms "substantially homogeneous", "substantially homogeneous form" and "substantial homogeneity" are used to indicate that the product is substantially free of by-products arising from unwanted polypeptide combinations (such as homomultimers).

[0126] Expressed in terms of purity, substantial homogeneity indicates that the amount of by-products does not exceed 10%, 9%, 8%, 7%, 6%, 4%, 3%, 2% or 1% by weight or is less than 1% by weight. In one realization, the by-product is less than 5%.

[0127] "Biological molecule" designates a nucleic acid, protein, carbohydrate, lipid and combinations thereof. In one embodiment, the biological molecule exists in nature.

[0128] "Isolated", when used to describe the various antibodies described herein, indicates an antibody that has been identified and separated and/or recovered from a cell or cell culture from which it was expressed. Contaminant components of its natural environment are materials that would typically interfere with diagnostic or therapeutic uses for the polypeptide, and may include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. In preferred embodiments, the antibody will be purified (1) to a degree sufficient to obtain at least fifteen residues of internal or N-terminal amino acid sequence using a centrifuge cup sequencer; or (2) to homogeneity by means of SDS-PAGE under reducing or non-reducing conditions, using Coomassie blue or, preferably, silver stains. Isolated antibody includes antibodies in situ in recombinant cells, as at least one component of the polypeptide natural environment will not be present. Ordinarily, however, isolated polypeptide will be prepared through at least one purification step.

[0129] By "linked" or "bonds", as used herein, is meant a direct peptide bond between a first and second amino acid sequence or a bond that involves a third amino acid sequence that is linked by peptide to the first and second amino acid sequences and between them. A linker peptide, for example, attached to the C-terminal end of one amino acid sequence and the N-terminal end of another amino acid sequence.

[0130] By "linker", as used herein, is meant a sequence of amino acids two or more amino acids in length. The linker can consist of non-polar or neutral polar amino acids. A linker may be, for example, 2 to 100 amino acids long, such as 2 to 50 amino acids long, for example 3, 5, 10, 15, 20, 25, 30, 35, 40, 45 or 50 amino acids of length. A ligand can be “divisible”, for example, through self-division or chemical or enzymatic division. Division sites in amino acid sequences and enzymes and substances that divide at these sites are well known in the art and also described herein.

[0131] By "teter", as used herein, is meant an amino acid linker that links two other amino acid sequences. A tether as described herein can link the N-terminus of an immunoglobulin heavy chain variable domain to the C-terminus of an immunoglobulin light chain constant domain. In specific embodiments, the teter is about 15 to 50 amino acids long, such as 20 to 26 amino acids long (e.g., 20, 21, 22, 23, 24, 25, or 26 amino acids long). A teter can be "dividable", for example, by means of self-division, chemical or enzymatic division using standard methods and reagents in the art. In certain specific embodiments, the teter comprises Gly-Gly-Ser repetitions.

[0132] Enzymatic division of a "linker" or "teter" may involve the use of an endopeptidase such as Lys-C, Asp-N, Arg-C, V8, Glu-C, chymotrypsin, trypsin, pepsin, papain , thrombin, Genenase, factor Xa, TEV (smoke corrosion virus cysteine protease), enterokinase, HRV C3 (human rhinovirus protease C3), Quininogenase, as well as subtilisin-like proprotein convertases (such as Furin (PC1) ), PC2 or PC3) or dibasic N-arginine convertase. Chemical division may involve the use of, for example, hydroxylamine, N-chlorosuccinimide, N-bromosuccinimide or cyanogen bromide.

[0133] "Endopeptidase Lys-C cleavage site", as used herein, is a lysine residue in an amino acid sequence that can be cleaved on the C-terminal side by endopeptidase Lys-C. Endopeptidase Lys-C cleaves on the C-terminal side of a lysine residue. In certain embodiments, the half-antibody further comprises a K222A mutation in the hinge region to remove the endogenous endopeptidase Lys-C cleavage site to preserve the structure of the half-antibody or bispecific antibody assembled upon teter cleavage by endopeptidase Lys- Ç.

[0134] "Joint region" is generally defined as extending from Glu216 to Pro230 of human IgG1 (Burton, Molec. Immunol. 22: 161-206 (1985)). The hinge regions of other IgG isotypes can be aligned with the IgG1 sequence by placing the first and last cysteine residues, forming interheavy chain S-S bonds at the same positions.

[0135] The "lower hinge region" of an Fc region is usually defined as the stretching of residues immediately C-terminal to the hinge region, that is, residues 233 to 239 of the Fc region. Prior to the present invention, FCYR binding was generally assigned to amino acid residues in the lower hinge region of an IgG Fc region.

[0136] The "CH2 domain" of a human IgG Fc region normally extends from about residues 231 to about 340 of IgG. The CH2 domain is unique in that it is not closely paired with another domain. On the other hand, two N-linked branched carbohydrate chains are interposed between the two CH2 domains of an intact native IgG molecule. It has been speculated that carbohydrate may provide a surrogate for domain-to-domain pairing and help stabilize the CH2 domain. Burton, Molec. Immunol. 22: 161-206 (1985).

[0137] The "CH3 domain" comprises the stretching of C-terminal residues to a CH2 domain in an Fc region (ie, from about amino acid residue 341 to about amino acid residue 447 of an IgG).

[0138] The term "Fc region" is used herein to define a C-terminal region of an immunoglobulin heavy chain, including native sequence Fc regions and variant Fc regions. Although the boundaries of the Fc region of an immunoglobulin heavy chain may vary, the human IgG heavy chain Fc region is usually defined as extending from an amino acid residue at position Cys226 or Pro230 to its carboxy terminus. . The C-terminal lysine (residue 447 according to the EU numbering system) of the Fc region can be removed, for example, during antibody production or purification or by recombinant elaboration of nucleic acid encoding an antibody heavy chain . Accordingly, an intact antibody composition can comprise antibody populations with all K447 residues removed, antibody populations without K447 residues removed, and antibody populations that have a mixture of antibodies with and without the K447 residue.

[0139] A "functional Fc region" has an "effector function" of a native sequence Fc region. Examples of "effector functions" include C1q binding, CDC, Fc receptor binding, ADCC, phagocytosis, down-regulation of cell surface receptors (such as B cell receptor; BCR) etc. Such effector functions generally require the Fc region to be combined with a binding domain (such as an antibody variable domain) and can be determined using various assays as described, for example, in the definitions herein.

[0140] A "native sequence Fc region" comprises an amino acid sequence identical to the amino acid sequence of an Fc region found in nature. Native sequence human Fc regions include a native sequence human IgG1 Fc region (allotypes A and non-A); native sequence human IgG2 Fc region; native sequence human IgG3 Fc region; and native sequence human IgG4 Fc region as well as naturally occurring variants thereof.

[0141] "Variant Fc region" comprises an amino acid sequence that differs from that of a native sequence Fc region by virtue of at least one amino acid modification, preferably one or more amino acid substitutions. Preferably, the variant Fc region contains at least one amino acid substitution compared to a native sequence Fc region or to the Fc region of a parent polypeptide, such as about one to about ten amino acid substitutions, and preferably about about one to about five amino acid substitutions in a native sequence Fc region or in the Fc region of the parent polypeptide. The variant Fc region of the present will preferably have at least about 80% homology to a native sequence Fc region and/or to an Fc region of a parent polypeptide, and more preferably, at least about 90% homology thereto. , more preferably at least about 95% homology.

[0142] "Fc complex", as used herein, designates two CH2 domains of an Fc region that interact with each other and/or two CH3 domains of an Fc region that interact with each other, in which the CH2 and/or domains the CH3 domains interact through bonds and/or forces (such as van der Waals, hydrophobic, and hydrophilic forces) that are not peptide bonds.

[0143] "Fc component", as used herein, designates a hinge region, a CH2 domain or a CH3 domain of an Fc region.

[0144] "CH Fc component" or "FcCH" as used herein means a polypeptide comprising a CH2 domain, CH3 domain or CH2 and CH3 domains of an Fc region.

[0145] Antibody "effector functions" designate the biological activities attributable to the Fc region (a native sequence Fc region or Fc region with amino acid sequence variation) of an antibody and vary according to the antibody's isotype. Examples of antibody effector functions include: complement dependent cytotoxicity and C1q binding; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; down-regulation of cell surface receptors (such as B cell receptors); and B cell activation.

[0146] In the present invention, “reducing condition” is defined based on the redox potential in a reaction (for example, in an assembly mixture) to indicate that the redox potential of the reaction is negative (-). The redox potential of the reaction under reducing conditions is preferably about -50 to -600 mV, -100 to -600 mV, -200 to -600 mV, -100 to -500 mV, -150 to -300 mV, most preferably about -300 to -500 mV and more preferably about -400 mV.

[0147] Any suitable method can be used to prepare a desired reducing condition. A desired reducing condition can be prepared, for example, by adding a reducing agent/reducing agent to the reaction (such as an assembly mixture in accordance with the present invention). Suitable reducers include, without limitation, dithiothreitol (DTT), tris(2-carboxyethyl)phosphine (TCEP), thioglycolic acid, ascorbic acid, thiol acetic acid, glutathione (GSH), betamercaptoethylamine, cysteine/cystine, GSH/glutathione disulfide ( GSSG), cysteamine/cystamine, glycylcysteine and betamercaptoethanol, preferably GSH. In certain specific embodiments, the reducer is a weak reducer which includes, without limitation, GSH, betamercaptoethylamine, cysteine/cystine, GSH/GSSG, cysteamine/cystamine, glycylcysteine and betamercaptoethanol, preferably GSH. In certain preferred embodiments, the reductant is GSH. It is within the ability of those of ordinary skill in the art to select suitable reducing agents under suitable concentrations and appropriate experimental conditions to achieve, in a reaction, the desired reducing condition. 10 mM L-reduced glutathione in a solution with a bispecific antibody protein concentration of 10 g/L at 20 °C, for example, will result in an initial redox potential of about -400 mV. Those skilled in the art can use any appropriate methods to measure the redox potential in a given reaction.

[0148] The reducing condition of the reaction can be estimated and measured using any appropriate method known in the art. The reducing condition can be measured, for example, using a resazurin indicator (blue to colorless discoloration under reducing conditions). For more accurate measurement, a redox potential meter (such as ORP electrode manufactured by BROADLEY JAMES®) can be used.

[0149] Alternatively, a reducing condition can be prepared by removing dissolved gases, especially dissolved oxygen, under reduced pressure of about 10 mmHg or less, preferably about 5 mmHg or less, more preferably about 3 mmHg or less, for about one to sixty minutes, preferably for about five to forty minutes.

[0150] In the present invention, it is preferred that reducing conditions are maintained immediately after mixing the first and second hinge-containing polypeptides (such as half-antibodies) throughout the assembly step. In certain embodiments, the reaction or assembly mixture is held at reducing conditions, preferably by about 50% or more, more preferably by about 70% or more, most preferably by about 90 % or more of reaction time. It is particularly preferred that the redox potential of the reaction medium be maintained at about -200 to -600 mV, more preferably -300 to -500 mV, preferably greater than about -400 mV, by about 50% or more, more preferably about 70% or more, most preferably about 90% or more of the reaction time.

[0151] In certain specific embodiments, the reducing condition is a weak reducing condition. The term "weak reducing" or "weakly reducing condition" as used herein means a reducing agent or reducing condition prepared by the reducing agent that has negative oxidation potential at 25°C. The oxidation potential of the reducer is preferably -50 to -600 mV, -100 to -600 mV, -200 to -600 mV, -100 to -500 mV, -150 to -300 mV, most preferably about - 300 to -500 mV, preferably greater than about -400 mV, where the pH is from 7 to 9 and the temperature is from 15°C to 39°C. Those skilled in the art will be able to select appropriate reducers to prepare a desired reduction condition. Those skilled in the art will recognize that a strong reductant, that is, one that has a more negative oxidation potential than the reductants mentioned above for the same concentration, pH and temperature, can be used at a lower concentration. In a preferred embodiment, the proteins will be capable of forming disulfide bonds in the presence of the reductant when incubated under the conditions indicated above. Examples of weak reducer include, without limitation, glutathione, betamercaptoethylamine, cystine/cysteine, GSH/GSSG, cysteamine/cystamine, glycylcysteine and betamercaptoethanol. In certain embodiments, oxidation potential similar to the 200X molar ratio of GSH:antibody can be used as a reference point for a weakly reducing condition in which efficient assembly using other reducers can be expected.

[0152] “Assembly mixture” is a solution comprising a first polypeptide that contains joint and a second polypeptide that contains joint. In certain embodiments, the assembly mixture is present in a reducing condition. In some embodiments, the assembly mixture is present in a weak reducing condition. In some other embodiments, the assembly mixture additionally comprises a weak reducer. The oxidation potential of the assembly mixture is -50 to -600 mV, -100 to -600 mV, -200 to -600 mV, -100 to -500 mV, -150 to -300 mV, from more preferably about -300 to -500 mV, more preferably about -400 mV, wherein the pH is from 7 to 9 and the temperature is from 15°C to 39°C.

[0153] Commercially available reagents indicated in the Examples were used in accordance with the manufacturer's instructions, unless otherwise indicated. The source of these cells identified in the examples below and throughout the descriptive report by ATCC accession numbers is the North American Collection of Type Crops, Manassas VA. Unless otherwise indicated, the present invention utilizes standard recombinant DNA technology procedures such as those described above and in the textbooks below: Sambrook et al, above; Ausubel et al, Current Protocols in Molecular Biology (Green Publishing Associates and Wiley Interscience, New York, 1989); Innis et al, PCR Protocols: A Guide to Methods and Applications (Academic Press, Inc., New York, 1990); Harlow et al, Antibodies: A Laboratory Manual (Cold Spring Harbor Press, Cold Spring Harbor, 1988); Gait, Oligonucleotide Synthesis (IRL Press, Oxford, 1984); Freshney, Animal Cell Culture, 1987; Coligan et al, Current Protocols in Immunology, 1991.

[0154] Throughout this specification and claims, the word "comprises" or variations such as "comprises" or "comprises" shall be understood to indicate the inclusion of an indicated integer or group of integers, but not the exclusion of any other whole number or group of whole numbers.

[0155] II. Construction of Heteromultimeric Proteins: Typically, the heteromultimeric proteins described herein will comprise a significant portion of an antibody Fc region. In other aspects, however, the heavy chain comprises only a part of the CH1, CH2 and/or CH3 domains.

[0156] Heteromultimerization domains: Heteromultimer proteins comprise a heteromultimerization domain. To generate a substantially homogeneous population of heterodimeric molecules, the heterodimerization domain must have a strong preference for forming heterodimers over homodimers. While the heteromultimeric proteins exemplified herein utilize knob-in-holes technology to facilitate heteromultimerization, those skilled in the art will appreciate other domains of heteromultimerization useful in the present invention.

[0157] Knobs-in-holes The use of knobs-in-holes as a method of producing multispecific antibodies is well known in the art. See U.S. Patent No. 5,731,168 issued March 24, 1998 and assigned to Genentech, PCT published No. WO 2009089004 published July 16, 2009 and assigned to Amgen and U.S. Patent published No. 20090182127 published 16 July 2009 and attributed to Novo Nordisk A/S. See also Marvin and Zhu, Acta Pharmacologica Sincia (2005) 26 (6): 649-658 and Kontermann (2005), Acta Pharmacol. Sin., 26: 1-9. A brief discussion is provided in the present.

[0158] "Knob" designates at least one amino acid side chain that projects from the interface of a first polypeptide and, therefore, can be positioned in a compensating hole at the adjacent interface (ie, the interface of a second polypeptide), in order to stabilize the heteromultimer and, in this way, favor the formation of heteromultimers over the formation of homomultimers, for example. The knob can exist at the original interface or can be introduced synthetically (by altering, for example, nucleic acid encoding the interface). Normally, nucleic acid encoding the interface of the first polypeptide is altered to encode the knob. To achieve this, nucleic acid encoding at least one "original" amino acid residue at the interface of the first polypeptide is replaced with nucleic acid encoding at least one "import" residue that has greater side chain volume than the original amino acid residue. . It will be appreciated that there may be more than one corresponding original and imported waste. The upper limit for the amount of original residues that are replaced is the total amount of residues at the interface of the first polypeptide. The side chain volumes of the various amino residues are shown in the table below. TABLE 1

a Peso molecular do aminoácido menos o da água. Valores de Handbook of Chemistry and Physics, 43a ed., Cleveland, Chemical Rubber Publishing Co., 1961. b Valores de A. A. Zamyatnin, Prog. Biophys. Mol. Biol. 24: 107123, 1972. c Valores de C. Chothia, J. Mol. Biol. 105: 1-14, 1975. A extensão acessível é definida nas Figuras 6 a 20 dessa referência.[0159] Properties of amino acid residues:

a Molecular weight of amino acid minus water. Values from Handbook of Chemistry and Physics, 43rd ed., Cleveland, Chemical Rubber Publishing Co., 1961. b Values from AA Zamyatnin, Prog. Biophys. Mol. Biol. 24: 107123, 1972. c Values from C. Chothia, J. Mol. Biol. 105: 1-14, 1975. The accessible range is defined in Figures 6 to 20 of this reference.

[0160] Preferred import residues for bulge (knob) formation are generally naturally occurring amino acid residues and are preferably selected from arginine (R), phenylalanine (F), tyrosine (Y) and tryptophan (W) . They are preferably superior tryptophan and tyrosine. In one embodiment, the original residue for formation of the knob (knob) has a small volume of side chain, such as alanine, asparagine, aspartic acid, glycine, serine, threonine, or valine. Examples of amino acid substitutions in the CH3 domain to form the knob include, without limitation, the T366W substitution.

[0161] "Hole" designates at least one amino acid side chain that is recessed from the interface of a second polypeptide and therefore accommodates a corresponding bulge (knob) over the adjacent interface of a first polypeptide. The hole can exist at the original interface or it can be introduced synthetically (by altering, for example, nucleic acid encoding the interface). Normally, nucleic acid encoding the interface of the second polypeptide is altered to encode the hole. To achieve this, the nucleic acid encoding at least one "original" amino acid residue at the interface of the second polypeptide is replaced with DNA encoding at least one "import" amino acid residue that has a side chain volume less than the amino acid residue. original. It will be appreciated that there may be more than one corresponding original and imported waste. The upper limit for the amount of original residues that are replaced is the total amount of residues at the interface of the second polypeptide. The side chain volumes of the various amino residues are shown in Table 1 above. Preferred imported residues for hole formation are typically naturally occurring amino acid residues and are preferably selected from alanine (A), serine (S), threonine (T) and valine (V). They are preferably superior serine, alanine or threonine. In one embodiment, the original residue for formation of the cavity (hole) has a large volume of side chain, such as tyrosine, arginine, phenylalanine or tryptophan. Examples of amino acid substitutions in the CH3 domain to generate the hole include, without limitation, the T366S, L368A, Y407A, Y407T and Y407V substitutions. In certain embodiments, the knob antibody half comprises T366W substitutions and the hole half antibody comprises T366S/L368A/Y407V substitutions.

[0162] "Original" amino acid residue is one that is replaced by an "import" residue that may have smaller or larger side chain volume than the original residue. The imported amino acid residue can be a naturally-occurring or non-naturally-occurring amino acid residue, but is preferably the former. "Natural-occurring" amino acid residues are those residues encoded by the genetic code and listed in Table 1 above. "Non-naturally occurring" amino acid residue means a residue which is not encoded by the genetic code, but which is capable of covalently joining adjacent amino acid residue(s) in the polypeptide chain. Examples of non-naturally occurring amino acid residues are norleucine, ornithine, norvaline, homoserine and other amino acid residue analogs such as those described in Ellman et al, Meth. Enzyme 202: 301-336 (1991), for example. To generate such non-naturally occurring amino acid residues, the procedures of Noren et al, Science 244: 182 (1989) and Ellman et al, above, can be used. Briefly, these procedures involve chemically activating a suppressor tRNA with a non-naturally occurring amino acid residue followed by RNA transcription and translation in vitro. The method according to the present invention involves the substitution of at least one original amino acid residue, but more than one original residue can be substituted. Typically, no more than the total residues at the interface of the first or second polypeptide will comprise original amino acid residues that are replaced. Typically, the original replacement waste is “buried”. By "buried" is meant that the residue is essentially inaccessible to the solvent. Generally, the imported residue is not cysteine, to avoid possible oxidation or mismatch of disulfide bonds.

[0163] The knob is "positionable" in the hole, which means that the spatial location of the knob and the hole (hole) on the interface of a first polypeptide and second polypeptide, respectively, and the knob and hole sizes are such that the knob can be located in the hole without significantly disturbing the normal association of the first and second polypeptides at the interface. As knobs such as Tyr, Phe and Trp typically do not extend perpendicularly from the interface axis and have preferred conformations, the alignment of a knob with a corresponding hole depends on the modeling of the protrusion pair ( knob) and hole (hole) based on a three-dimensional structure, such as that obtained by means of X-ray crystallography or nuclear magnetic resonance (NMR). This can be achieved using methods widely accepted in the art.

[0164] By "original or template nucleic acid" is meant the nucleic acid encoding a polypeptide of interest that can be "altered" (ie, genetically engineered or mutated) to encode bulge (knob) or cavity ( hole). The original or starting nucleic acid can be a naturally occurring nucleic acid or can comprise a nucleic acid that has undergone prior alteration (e.g., a humanized antibody fragment). By "altering" the nucleic acid, it is meant that the original nucleic acid is mutated through insertion, deletion or substitution of at least one codon encoding an amino acid residue of interest. Typically, a codon encoding an original residue is replaced by a codon encoding an imported residue. Methods of genetically modifying DNA in this way have been discussed in Mutagenesis: a Practical Approach, MJ McPherson, Ed., (IRL Press, Oxford, UK (1991)) and include, for example, site-directed mutagenesis, cluster mutagenesis and polymerase chain reaction (PCR) mutagenesis. Upon mutation of the parent/template nucleic acid, a parent/template polypeptide encoded by the parent/template nucleic acid is therefore altered accordingly.

[0165] The bulge (knob) or cavity (hole) can be "introduced" at the interface of a first or second polypeptide by synthetic means, such as by recombinant methods, in vitro peptide synthesis, in which these methods are intended to introducing previously described non-naturally occurring amino acid residues by means of enzymatic or chemical peptide coupling or some combination of these methods. Consequently, the knob or hole that is "introduced" is "non-naturally occurring" or "non-native", meaning that it does not exist in nature or in the parent polypeptide (such as a humanized monoclonal antibody) .

[0166] Generally, the amino acid residue imported for knob formation has a relatively small amount of "rotamers" (such as about 3 to 6). "Rotamer" is an energetically favorable conformation of an amino acid side chain. The amount of rotamers from the various amino acid residues is discussed in Ponders and Richards, J. Mol. Biol. 193: 775-791 (1987).

[0167] III. Vectors, host cells and recombinant methods: For the recombinant production of a heteromultimeric protein (such as a bispecific antibody) in accordance with the present invention, the nucleic acid encoding it is isolated and inserted into a reproducible vector for further cloning (amplification of the DNA) or for expression. The antibody-encoding DNA is readily isolated and sequenced using conventional procedures (through, for example, the use of oligonucleotide probes that are capable of specifically binding to genes encoding the antibody heavy and light chains). Several vectors are available. The vector selected depends in part on the host cell to be used. Generally, preferred host cells are of prokaryotic or eukaryotic origin (generally from mammals, but also including fungi (such as yeast), insects, plants, and nucleated cells from other multicellular organisms). It will be appreciated that constant regions of any isotype can be used for this purpose, including constant regions from IgG, IgM, IgA, IgD and IgE, and that such constant regions can be obtained from any human or animal species. In certain embodiments, the constant region is IgG, particularly IgG1, IgG2 or IgG4.

[0168] A host cell is constructed in such a way that it expresses a hinge-containing polypeptide and comprises a heterodimerization domain wherein the host cell does not express a hinge-containing polypeptide and comprises a second heterodimerization domain. The. Generation of Heteromultimeric Proteins Using Prokaryotic Host Cells: 1. Construction of Vectors: Polynucleotide sequences encoding polypeptide components of the heteromultimeric proteins (such as antibodies) according to the present invention can be obtained using standard recombinant methods. Desired polynucleotide sequences can be isolated and sequenced, for example, from antibody-producing cells, such as hybridoma cells. Alternatively, polynucleotides can be synthesized using a nucleotide synthesizer or PCR techniques. Once obtained, the sequences encoding the polypeptides are inserted into a recombinant vector capable of reproducing and expressing heterologous polynucleotides in prokaryotic hosts. Many vectors that are available and known in the art can be used for the purpose of the present invention. Selection of an appropriate vector will depend primarily on the size of the nucleic acids to be inserted into the vector and the specific host cell to be transformed with the vector. Each vector contains several components, depending on its function (amplification or expression of heterologous polynucleotide, or both) and its compatibility with the specific host cell in which it resides. Vector components generally include, but are not limited to: origin of reproduction, selectable marker gene, promoter, ribosome binding site (RBS), signal sequence, heterologous nucleic acid insert, and transcription termination sequence.

[0169] Generally, plasmid vectors that contain control and replication sequences that are derived from species compatible with the host cell are used with respect to those hosts. The vector normally carries a breeding site as well as marker sequences that are capable of providing phenotypic selection in transformed cells. E. coli, for example, is typically transformed using pBR322, a plasmid derived from an E. coli species. pBR322 contains genes encoding ampicillin (Amp) and tetracycline (Tet) resistance and therefore provides an easy means of identifying transformed cells. pBR322, its derivatives or other microbial plasmids or bacteriophage may also contain or be modified so that they contain promoters which can be used by the microbial organism for the expression of endogenous proteins. Examples of pBR322 derivatives used for the expression of specific antibodies are described in detail in Carter et al, U.S. Patent No. 5,648,237.

[0170] In addition, phage vectors that contain control and replica sequences that are compatible with the host microorganism can be used as transformation vectors with respect to these hosts. Bacteriophages such as XGEM.TM.-11, for example, can be used in the manufacture of recombinant vectors that can be used to transform susceptible host cells such as E. coli LE392.

[0171] The expression vector according to the present invention may comprise two or more pairs of promoter and cistron, which encode each of the components of polypeptides. Promoter is an untranslated regulatory sequence located upstream (5') for a cistron that modulates its expression. Prokaryotic promoters typically fall into two classes, inducible and constitutive. The inducible promoter is a promoter that initiates higher levels of transcription from the cistron under its control in response to changes in growing conditions, such as the presence or absence of nutrients or a change in temperature.

[0172] A large amount of promoters recognized by a number of potential host cells is well known. The selected promoter can be operably linked to cistron DNA encoding, for example, the light or heavy chain by removing the promoter from the source DNA by restriction enzyme digestion and inserting the isolated promoter sequence into the vector in accordance with the present invention. Both the native promoter sequence and many heterologous promoters can be used to direct amplification and/or expression of target genes. In some embodiments, heterologous promoters are used as they generally allow for greater transcription and higher yields of the expressed target gene compared to the native target polypeptide promoter.

[0173] Suitable promoters for use with prokaryotic hosts include the PhoA promoter, the lactose and β-galactamase promoter systems, the tryptophan (trp) promoter system, and hybrid promoters such as the tac or trc promoter. However, other promoters that are functional in bacteria (such as other known bacterial or phage promoters) are also suitable. Their nucleotide sequences have been published to allow those skilled in the art to operably link them to cistrons that encode heteromultimeric protein genes, such as target light and heavy chains (Siebenlist et al (1980), Cell, 20: 269 ) using linkers or adapters to provide any necessary restriction sites.

[0174] In one aspect of the present invention, each cistron within the recombinant vector comprises a secretion signal sequence component that directs the translocation of the expressed polypeptides across a membrane. Generally speaking, the signal sequence can be a component of the vector or it can be a part of the target polypeptide DNA that is inserted into the vector. The signal sequence selected for the purpose of the present invention must be recognized and processed (i.e., split by signal peptidase) by the host cell. For prokaryotic host cells that do not recognize and process the native signal sequences for the heterologous polypeptides, the signal sequence is replaced by a prokaryotic signal sequence selected, for example, from the group consisting of alkaline phosphatase, penicillinase, lpp or heat-stable enterotoxin II (STII), LamB, PhoE, PelB, OmpA and MBP leaders. In one embodiment of the present invention, the signal sequences used in the two cistrons of the expression system are STII signal sequences or variants thereof.

[0175] In another aspect, the production of immunoglobulins according to the present invention can occur in the cytoplasm of the host cell and, therefore, does not require the presence of secretion signal sequences in each cistron. In this regard, immunoglobulin light and heavy chains are expressed, folded and assembled to form functional immunoglobulins within the cytoplasm. Certain host strains (such as the E. coli trxB- strains) provide cytoplasmic conditions that are favorable for the formation of disulfide bonds to allow for proper folding and assembly of the expressed protein subunits. See Proba and Pluckthun, Gene, 159: 203 (1995).

Suitable prokaryotic host cells for the expression of heteromultimeric proteins (such as antibodies) in accordance with the present invention include archaebacteria and eubacteria, such as gram-negative or gram-positive organisms. Examples of useful bacteria include Escherichia (such as E. coli), Bacilli (such as B. subtilis), Enterobacteria, Pseudomonas species (such as P. aeruginosa), Salmonella typhimurium, Serratia marcescans, Klebisella, Proteus, Shigella, Rhizobia, Vitreoscilla or Paracoccus. In one embodiment, gram-negative cells are used. In one embodiment, E. coli cells are used as hosts for the present invention. Examples of E. coli strains include strain W3110 (Bachmann, Cellular and Molecular Biology, vol. 2 (Washington DC, American Society for Microbiology, 1987), pp. 1190-1219; ATCC Deposit No. 27325) and its derivatives , including strain 33D3 that has genotype W3110 ΔfhuA (ΔtonA) ptr3 lac lq lacL8 ΔompTΔ (nmpc-fepE) degP41 kanR (US Patent No. 5,639,635). Other strains and their derivatives, such as E. coli 294 (ATCC 31,446), E. coli B, E. coli 1776 (ATCC 31,537), and E. coli RV308 (ATCC 31,608) are also suitable. In one embodiment, E. coli Δlpp finds specific use. These examples are illustrative and not limiting. Methods of constructing derivatives of any of the aforementioned bacteria that have defined genotypes are known in the art and described, for example, in Bass et al, Proteins, 8: 309-314 (1990). It is usually necessary to select the appropriate bacteria considering the replica's ability to reproduce in the bacteria cells. E. coli, Serratia or Salmonella species, for example, can suitably be used as a host by employing well known plasmids such as pBR322, pBR325, pACYC177 or pKN410 to provide the replica. Typically, the host cell will secrete minimal amounts of proteolytic enzymes and additional protease inhibitors may desirably be incorporated into the cell culture.

[0177] ii. Polypeptide Production: Host cells are transformed with the expression vectors described above and cultured in conventional nutrient media modified as appropriate for inducing promoters, selecting transformants, or amplifying the genes encoding the desired sequences.

[0178] Transformation indicates the introduction of DNA into the prokaryotic host, in such a way that the DNA is reproducible, either in the form of extrachromosomal element or by chromosomal integrant. Depending on the host cell used, transformation is performed using standard techniques appropriate to those cells. Calcium treatment employing calcium chloride is generally used for bacterial cells that contain substantial cell wall barriers. Another transformation method employs polyethylene glycol/DMSO. Yet another commonly used method is electroporation.

[0179] The prokaryotic cells used to produce the polypeptides according to the present invention are cultivated in media known in the art and suitable for culturing the selected host cells. Examples of suitable media include Luria broth (LB) plus necessary nutrient supplements. In some embodiments, the media also contains a selection agent, selected based on the construction of the expression vector, to selectively allow the growth of prokaryotic cells that contain the expression vector. Ampicillin is added to the media, for example, to grow cells expressing the ampicillin resistant gene.

[0180] Any necessary supplements other than carbon, nitrogen and inorganic phosphate sources may also be included in appropriate concentrations introduced alone or in the form of a mixture with another supplement or medium, such as a complex nitrogen source. Optionally, the culture medium can contain one or more reducing agents selected from the group consisting of glutathione, cysteine, cystamine, thioglycolate, dithioerythritol and dithiothreitol.

[0181] Prokaryotic host cells are cultured under appropriate temperatures. For E. coli growth, for example, the preferred temperature ranges from about 20°C to about 39°C, more preferably about 25°C to about 37°C, most preferably about 30°C Ç. The pH of the medium can be any pH ranging from about 5 to about 9, depending primarily on the host organism. For E. coli, the pH is preferably from about 6.8 to about 7.4, and most preferably about 7.0.

[0182] If an inducible promoter is used in the expression vector according to the present invention, the expression of proteins is induced under appropriate conditions for the activation of the promoter. In one aspect of the present invention, PhoA promoters are used to control the transcription of polypeptides. Consequently, the transformed host cells are cultured in phosphate-limiting medium for induction. Preferably, the phosphate-limiting medium is the C.R.A.P. medium (see, for example, Simmons et al, J. Immunol. Methods (2002), 263: 133-147). A number of other inducers can be used, in accordance with the vector construction employed, as is known in the art.

[0183] In one embodiment, the first and second hinge-containing host cells are cultured separately and the polypeptides expressed in accordance with the present invention are secreted and recovered from the periplasm of the host cells separately. In a second embodiment, the first and second hinge-containing host cells are cultured separately and, prior to isolation of the hinge-containing polypeptides, the two host cell cultures are mixed together and the cells are pelleted. In a third embodiment, the first and second hinge-containing host cells are separately cultured, centrifuged, re-suspended separately, and then mixed prior to isolation of the hinge-containing polypeptides. In a fourth embodiment, the first and second hinge-containing host cells are grown together in the same culture vessel. Protein recovery typically involves disruption of the microorganism's cell membrane, usually by means such as osmotic shock, sonication or lysis. After cell disruption, cell debris or whole cells can be removed by centrifugation or filtration. Proteins can be further purified, for example, by means of affinity resin chromatography. Alternatively, proteins can be transported or secreted into the culture medium and isolated therein. Recombinant proteins expressed with an exogenous sequence tag (or epitope tag) can facilitate the purification step. The method of cloning and purifying proteins that contain an exogenous sequence tag (including, without limitation, the His tag and the GST tag) is well known in the art. Cells can be removed from the culture and the culture supernatant is filtered and concentrated for further purification of the produced proteins. Expressed polypeptides can be further isolated and identified using commonly known methods such as polyacrylamide gel electrophoresis (PAGE) and Western Blot testing. Isolated polypeptides will be used to produce heteromultimeric proteins.

[0184] In one aspect of the present invention, the production of heteromultimeric proteins (such as antibodies) is conducted in large quantity through a fermentation process. Several large-scale batch feed fermentation procedures are available for the production of recombinant proteins. Large scale fermentations have a capacity of at least one thousand liters, preferably around 1000 to 100,000 liters of capacity. These fermenters use agitator boosters to deliver oxygen and nutrients, especially glucose (the preferred carbon and energy source). Small scale fermentation generally refers to fermentation in a fermenter that has a volumetric capacity of no more than about a hundred liters and can range from about a liter to about a hundred liters.

[0185] In a fermentation process, induction of protein expression is typically initiated after culturing the cells under appropriate conditions to desired density, such as OD550 of about 180 to 220, stage at which the cells are in the phase initial stationary. A number of inducers can be used, in accordance with the vector construction employed, as is known in the art and described above. Cells can be cultured for shorter periods before induction. Cells are normally induced for about twelve to fifty hours, although a longer or shorter induction time can be used.

[0186] To increase the production yield and quality of the polypeptides according to the present invention, various fermentation conditions can be modified. To improve proper assembly and folding of secreted heteromultimeric proteins (such as antibodies), for example, additional vectors that overexpress chaperone proteins, such as Dsb proteins (DsbA, DsbB, DsbC, DsbD and/or DsbG) or FkpA ( peptidylprolyl cis, trans-isomerase with chaperone activity), can be used to co-transform prokaryotic host cells. Chaperone proteins have been shown to facilitate the proper folding and solubility of heterologous proteins produced in bacterial host cells. Chen et al (1999), J. Bio. Chem. 274: 19601-19605; Georgiou et al, U.S. Patent No. 6,083,715; Georgiou et al, U.S. Patent No. 6,027,888; Bothmann and Pluckthun (2000), J. Biol. Chem. 275: 17100-17105; Ramm and Pluckthun (2000), J. Biol. Chem. 275: 17106-17113; Arie et al (2001), Mol. Microbiol. 39:199-210.

[0187] To minimize proteolysis of expressed heterologous proteins (especially those that are proteolytically sensitive), certain host strains deficient in proteolytic enzymes can be used for the present invention. Host cell lines can be modified, for example, to effect genetic mutation(s) in genes encoding known bacterial proteases, such as Protease III, OmpT, DegP, Tsp, Protease I, Protease Mi, Protease V, Protease VI and its combinations. Some E. coli protease-deficient strains are available and described, for example, in Joly et al (1998), Proc. Natl. Academic Sci. U.S.A. 95: 2773-2777; Georgiou et al, U.S. Patent No. 5,264,365; Georgiou et al, U.S. Patent No. 5,508,192; Hara et al, Microbial Drug Resistance, 2:63-72 (1996).

[0188] In one embodiment, E. coli strains deficient in proteolytic enzymes and transformed with plasmids that over-express one or more chaperone proteins are used as host cells in the expression system according to the present invention. In a second embodiment, the E. coli strain is deficient in an outer membrane lipoprotein (Δlpp).

[0189] iii. Purification of heteromultimeric proteins:

[0190] In one embodiment, the heteromultimeric protein produced herein is further purified to obtain preparations that are substantially homogeneous for further testing and uses. Standard protein purification methods known in the art can be employed. The following procedures are examples of appropriate purification procedures: fractionation over immunoaffinity or ion exchange columns; ethanol precipitation; reverse phase HPLC; chromatography on silica or on an ion exchange resin such as DEAE; chromatoconcentration; SDS-PAGE, ammonium sulphate precipitation and gel filtration using eg Sephadex G-75.

[0191] In one aspect, Protein A immobilized on solid phase is used for immunoaffinity purification, for example, of the products of full-length or half-antibody antibodies according to the present invention. Protein A is a 41 kD cell wall protein from Staphylococcus aureus that binds with high affinity to the Fc region of antibodies. Lindmark et al (1983), J. Immunol. Meth. 62: 1-13. The solid phase to which Protein A is immobilized is preferably a column comprising a surface of glass or silica, more preferably a glass column with controlled pores or column of silicic acid. In some applications, the column has been coated with a reagent, such as glycerol, to try to prevent non-specific adherence of contaminants.

[0192] As a first purification step, the preparation derived from cell culture as described above is applied onto the solid phase immobilized with Protein A to allow specific binding of the antibody of interest to Protein A. The solid phase is then washed to remove contaminants non-specifically bound to the solid phase. Heteromultimeric protein (such as antibody) is recovered from the solid phase through elution.

[0193] b. Generation of Heteromultimeric Proteins Using Eukaryotic Host Cells: Vector components generally include, but are not limited to, one or more of the following: signal sequence, origin of reproduction, one or more marker genes, amplifier element, promoter, and transcription termination sequence .

[0194] i. Signal Sequence Component: A vector for use in eukaryotic host cells may also contain a signal sequence or other polypeptide that has a specific cleavage site at the N-terminus of the mature protein or polypeptide of interest. The preferably selected heterologous signal sequence is recognized and processed (i.e., split by signal peptidase) by the host cell. In mammalian cell expression, mammalian signal sequences as well as viral secretion leaders, such as the herpes simplex gD signal, are available. The DNA for that precursor region is ligated in reading frame to DNA encoding the desired heteromultimeric protein(s) (such as antibodies).

[0195] ii. Origin of reproduction: Generally, an origin of reproduction component is not required for mammalian expression vectors. The SV40 origin, for example, can typically be used, but only because it contains the early promoter.

[0196] iii. Selection gene component: Cloning and expression vectors may contain a selection gene, also called a selectable marker. Typical selection genes encode proteins that (a) confer resistance to antibiotics or other toxins, such as ampicillin, neomycin, methotrexate or tetracycline; (b) complement auxotrophic deficiencies, when relevant; or (c) provide key nutrients not available from complex media.

[0197] An example selection scheme uses a drug to interrupt the growth of host cells. Cells that are successfully transformed with heterologous genes produce proteins that confer drug resistance and thus survive the selection regimen. Examples of this dominant selection use the drugs neomycin, mycophenolic acid and hygromycin.

[0198] Another example of suitable selectable markers for mammalian cells are those that allow the identification of cells competent to take up antibody nucleic acid, such as DHFR, thymidino kinase, metallothionein I and II, preferably primate metallothionein genes, adenosine deaminase, ornithine decarboxylase etc.

[0199] Cells transformed with the DHFR selection gene, for example, are first identified by growing all the transformants in a culture medium that contains methotrexate (Mtx), a concurrent antagonist of DHFR. An appropriate host cell, when employing wild-type DHFR, is the Chinese hamster ovary (CHO) cell line deficient in DHFR activity (eg ATCC CRL-9096).

[0200] Alternatively, host cells (particularly wild-type hosts that contain endogenous DHFR) transformed or co-transformed with DNA sequences encoding an antibody, wild-type DHFR protein and other selectable marker such as aminoglycoside 3'-phosphotransferase (APH) can be selected by means of cell growth in a medium which contains a selection agent for the selectable marker, such as an aminoglycoside antibiotic, for example kanamycin, neomycin or G418. See, for example, U.S. Patent No. 4,965,199.

[0201] iv. Promoter component: Cloning and expression vectors typically contain a promoter that is recognized by the host organism and is operably linked to the nucleic acid of the hinge-containing polypeptide(s) (eg, antibody). Promoter sequences for eukaryotes are known. Virtually all eukaryotic genes contain an AT-rich region located about 25 to 30 bases above where transcription begins. Another sequence found 70 to 80 bases above the start of transcription of many genes is a CNCAAT region, where N can be any nucleotide. At the 3' end of most eukaryotic genes is an AATAAA sequence which may be the addition signal from the poly A end to the 3' end of the coding sequence. All these sequences are properly inserted into eukaryotic expression vectors.

[0202] The transcription of polypeptide(s) that contain(s) linkage (eg, half-antibody) from vectors in mammalian host cells is controlled, for example, by promoters obtained from the genomes of such viruses. such as polyoma virus, chickenpox virus, adenovirus (such as Adenovirus 2), bovine papilloma virus, avian sarcoma virus, cytomegalovirus, retrovirus, hepatitis B virus and Simian Virus 40 (SV40), from heterologous mammalian promoters, such as such as the actin promoter or immunoglobulin promoter, or heat shock promoters, provided such promoters are compatible with host cell systems.

[0203] The early and late promoters of the SV40 virus are conveniently obtained in the form of a SV40 restriction fragment which also contains the viral origin of reproduction of SV40. The human cytomegalovirus immediate early promoter is conveniently obtained in the form of a HindIII E restriction fragment. A DNA expression system in mammalian hosts using bovine papilloma virus as a vector is described in U.S. Patent No. 4,419,446 . A modification of this system is described in U.S. Patent No. 4,601,978. See also Reyes et al, Nature 297: 598-601 (1982) on the expression of human β-interferon cDNA in mouse cells under the control of a herpes simplex virus thymidine kinase promoter. Alternatively, the Rous Sarcoma Virus long terminal repeat can be used as a promoter.

[0204] v. Amplifier element component: The transcription of DNA encoding the polypeptide(s) containing the link (such as antibody) by higher eukaryotes can be increased by inserting an amplifier sequence into the vector. Many enhancer sequences are now known from mammalian genes (eg, globin, elastase, albumin, a-fetoprotein and insulin genes). In addition, an enhancer of a eukaryotic cell virus can be used. Examples include the SV40 posterior side of the reproduction origin enhancer (bp 100-270), the cytomegalovirus early promoter enhancer, the polyoma posterior side of the reproduction origin enhancer, and adenovirus enhancers. See also Yaniv, Nature 297: 17-18 (1982) for a description of elements that enhance the activation of eukaryotic promoters. The enhancer can be split in the vector 5' or 3' to the antibody polypeptide coding sequence, provided amplification is achieved, but is generally located 5' to the promoter.

[0205] vi. Transcription Termination Component: Expression vectors used in eukaryotic host cells will also typically contain sequences necessary for transcription termination and mRNA stabilization. These sequences are usually available from the 5' and occasionally 3' untranslated regions of viral or eukaryotic DNAs or cDNAs. These regions contain nucleotide segments transcribed as polyadenylated fragments in the untranslated portion of the antibody-encoding mRNA. A useful transcription termination component is the bovine growth hormone polyadenylation region. See WO 94/11026 and the expression vector described therein.

[0206] vii. Selection and Transformation of Host Cells: Appropriate host cells for cloning or expressing the DNA in the vectors herein include the higher eukaryotic cells described herein, including vertebrate host cells. The propagation of vertebrate cells in culture (tissue culture) has become a routine procedure. Examples of useful mammalian host cell lines are monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney lineage (293 or 293 cells subcloned for growth in suspension culture, Graham et al, J. Gen. Virol. 36:59 (1977)); baby hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells/-DHFR (CHO, Urlaub et al, Proc. Natl. Acad. Sci. U.S.A. 77: 4216 (1980)); mouse sertoli cells (TM4, Mather, Biol. Reprod. 23: 243-251 (1980)); monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL 2), canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al, Annals N.Y. Acad. Sci. 383: 44-68 (1982)); MRC 5 cells; FS4 cells; and human hepatoma lineage (Hep G2).

[0207] Host cells are transformed with the cloning or expression vectors described above for the production of polypeptide(s) that contain(s) the desired articulation(s) (for example, antibody) and cultured in conventional nutrient media modified as per suitable for inducing promoters, selecting transformers or amplifying the genes encoding the desired sequences.

[0208] viii. Cultivation of Host Cells: Host cells used for the production of polypeptide(s) containing the desired linkage(s) (eg, antibody) in accordance with the present invention may be cultured in a variety of media. Commercially available media such as Ham's F10 (Sigma), Minimal Essential Medium (MEM, Sigma), RPMI-1640 (Sigma) and Dulbecco's Modified Eagle's Medium (DMEM, Sigma) are suitable for culturing the host cells. Furthermore, any of the means described in Ham et al, Meth. Enz. 58: 44 (1979), Barnes et al, Anal. Biochem. 102: 255 (1980), U.S. Patent Nos. 4,767,704, 4,657,866, 4,927,762, 4,560,655, or 5,122,469; WO 90/03430; WO 87/00195; or U.S. Patent Ref. 30,985 can be used as a culture medium for the host cells. Any of these media can be supplemented as needed with hormones and/or other growth factors (such as insulin, transferrin or epidermal growth factor), salts (such as sodium chloride, calcium, magnesium and phosphate), buffers (such as such as HEPES), nucleotides (such as adenosine and thymidine), antibiotics (such as the drug GENTAMYCIN®), trace elements (defined as inorganic compounds normally present in final concentrations in the micromolar range), and glucose or equivalent energy source. Any other necessary supplements may also be included at appropriate concentrations that would be known to those skilled in the art. Cultivation conditions, such as temperature, pH, and the like, are those previously used with the host cell selected for expression and will be apparent to those of ordinary skill in the art.

[0209] ix. Purification of heteromultimeric proteins: Using recombinant techniques, hinge-containing polypeptides can be produced intracellularly or directly secreted into the medium. If the joint-containing polypeptide is produced intracellularly, as a first step, the particulate fragments, whether they are host cells or lysed fragments, are removed, for example, by means of centrifugation or ultrafiltration. When the hinge-containing polypeptide is secreted into the medium, supernatants from such expression systems are generally concentrated first using a commercially available protein concentration filter, such as an Amicon or Millipore Pellicon ultrafiltration unit. Protease inhibitor, such as PMSF, can be included in any of the above steps to inhibit proteolysis and antibiotics can be included to prevent the growth of unanticipated contaminants.

[0210] The heteromultimeric composition prepared from the cells can be purified using, for example, ion exchange, hydrophobic interaction chromatography, hydroxylapatite chromatography, gel electrophoresis, dialysis and affinity chromatography, in which affinity chromatography is the technique of preferred purification. Combinations of the methods mentioned above are also contemplated. The suitability of protein A as an affinity binder depends on the species and isotype of any immunoglobulin Fc domain that is present in the antibody. Protein A can be used to purify antibodies that are based on human y1, Y2 or Y4 heavy chains (Lindmark et al, J. Immunol. Meth. 62: 1-13 (1983)). Protein G is recommended for all mouse isotypes and for human Y3 (Guss et al, 1986, EMBO J. 5:1567-1575 (1986)). The matrix to which the affinity ligand is attached is most often agarose, but other matrices are available. Mechanically stable matrices, such as controlled pore glass or poly(styrenedivinyl)benzene, allow for higher flow rates and shorter processing times than can be achieved with agarose. When the antibody comprises a CH3 domain, Bakerbond ABX® resin (J.T. Baker, Phillipsburg NJ) is useful for purification. Other protein purification techniques such as fractionation on an ion exchange column, ethanol precipitation, Reverse Phase HPLC, silica chromatography, chromatography on SEPHAROSE® heparin, chromatography on an anion or cation exchange resin (such as a polyaspartic acid column), chromatoconcentration, SDS-PAGE and ammonium sulfate precipitation are also available, depending on the antibody to be recovered.

[0211] After any preliminary purification step(s), the mixture comprising the antibody of interest and contaminants can be subjected to hydrophobic interaction chromatography under low pH, using elution buffer at pH about from 2.5 to 4.5, preferably carried out under low salt concentrations (eg about 0 to 0.25 M salt). Production of the heteromultimeric proteins may comprise, alternatively or in addition (to any of the specific methods above), dialysis of a solution comprising a mixture of the polypeptides.

[0212] x. Antibody production using Bacilloviruses: Recombinant bacilloviruses can be generated by co-transfecting a plasmid encoding an antibody or antibody fragment and BaculoGold® virus DNA (Pharmingen) into an insect cell such as a Spodoptera frugiperda cell (such as Sf9 cells; ATCC CRL 1711) or a Drosophila melanogaster S2 cell, using, for example, lipofectin (commercially available from GIBCO-BRL). In a specific example, an antibody sequence is fused up in the stream from an epitope tag contained in a bacillovirus expression vector. These epitope tags include poly-His tags. A variety of plasmids can be employed, which include plasmids derived from commercially available plasmids such as pVL1393 (Novagen) or pAcGP67B (Pharmingen). Briefly, the sequence encoding an antibody or its fragment can be amplified through PCR with primers complementary to the 5' and 3' regions. The 5' primer can incorporate flanking (selected) restriction enzyme sites. The product can then be digested with selected restriction enzymes and subcloned into the expression vector.

[0213] After transfection with the expression vector, host cells (such as Sf9 cells) are incubated for four to five days at 28 °C and the released virus is harvested and used for further amplifications. Viral infection and protein expression can be performed as described, for example, by O'Reilley et al (Baculovirus expression vectors: A Laboratory Manual, Oxford: Oxford University Press (1994)).

[0214] Expressed poly-His tagged antibody can then be purified, for example, by Ni2+ chelate affinity chromatography as follows. Extracts can be prepared from recombinant virus-infected Sf9 cells as described by Rupert et al (Nature 362: 175-179 (1993)). Briefly, Sf9 cells are washed, resuspended in sonication buffer (25 ml HEPES, pH 7.9; 12.5 mM MgCl2; 0.1 mM EDTA; 10% glycerol; 0.1% NP-40; 0.4 M KCl) and sonicated twice for twenty seconds on ice. The sonicates are released by centrifugation and the supernatant is diluted fifty times in loading buffer (50 mM phosphate, 300 mM NaCl, 10% glycerol, pH 7.8) and filtered through a 0.45 filter µm. Ni2+-NTA agarose column (commercially available from Qiagen) is prepared with bed volume of 5 ml, washed with 25 ml of water and equilibrated with 25 ml of loading buffer. The filtered cell extract is loaded onto the column at 0.5 ml per minute. The column is washed to baseline A280 with loading buffer, at which point fraction collection begins. The column is then washed with secondary wash buffer (50 mM phosphate; 300 mM NaCl, 10% glycerol, pH 6.0), which elutes non-specifically bound protein. After reaching A280 baseline again, the column is developed with a 0 to 500 mM gradient of imidazole in the secondary wash buffer. 1 ml fractions are collected and analyzed by means of SDS-PAGE and silver stains or Western Blot with Ni2+-NTA conjugated to alkaline phosphatase (Qiagen). Fractions containing the eluted His10-tagged antibody are pooled and dialyzed against loading buffer.

[0215] Alternatively, antibody purification can be performed using known chromatography techniques, which include, for example, Protein A or Protein G column chromatography. In one embodiment, the antibody of interest can be recovered from the solid phase of the column. by elution into a solution containing a chaotropic agent or mild detergent. Examples of chaotropic agents and mild detergents include, but are not limited to, guanidine-HCl, urea, lithium perchlorate, arginine, histidine, SDS (sodium dodecyl sulfate), Tween, Triton and NP-40, all of which are commercially available.

[0216] iv. Target molecules: Examples of molecules that can be targeted by a heteromultimeric protein in accordance with the present invention include, but are not limited to, soluble serum proteins and their receptors and other membrane-bound proteins (such as adhesins). See WO 2011/133886, which is incorporated herein by reference in its entirety.

[0217] In another embodiment, the heteromultimeric protein according to the present invention is capable of binding one, two or more cytokines, cytokine-related proteins and cytokine receptors selected from the group consisting of BMPI, BMP2, BMP3B (GDFIO ), BMP4, BMP6, BMP8, CSFI (M-CSF), CSF2 (GM-CSF), CSF3 (G-CSF), EPO, FGFI (aFGF), FGF2 (bFGF), FGF3 (int-2), FGF4 ( HST), FGF5, FGF6 (HST-2), FGF7 (KGF), FGF9, FGF10, FGF11, FGF12, FGF12B, FGF14, FGF16, FGF17, FGF19, FGF20, FGF21, FGF23, IGF1 , IGF2, IFNAI, IFNA2, IFNA4 , IFNA5, IFNA6, IFNA7, IFNBI, IFNG, IFNWI, FELI, FELI (EPSELON), FELI (ZETA), IL1A, IL1B, IL2, IL3, IL4, IL5, IL6, IL7, IL8, IL9, IL10, IL11, IL12A , IL12B, IL13, IL14, IL15, IL16, IL17, IL17B, IL18, IL19, IL20, IL22, IL23, IL24, IL25, IL26, IL27, IL28A, IL28B, IL29, IL30, PDGFA, PDGFB, TGFA, TGFB1, TGFB2 , TGFB3, LTA (TNF-b), LTB, TNF (TNF-a), TNFSF4 (OX40 linker), TNFSF5 (CD40 linker), TNFSF6 (FasL), TNFSF7 (CD27 linker), TNFSF8 (CD30 linker), TNFSF9 ( read gante 4-1 BB), TNFSFIO (TRAIL), TNFSF1 I (TRANCE), TNFSF12 (APO3L), TNFSF13 (April), TNFSF13B, TNFSF14 (HVEM-L), TNFSF15 (VEGI), TNFSF18, HGF (VEGFD), VEGF , VEGFB, VEGFC, ILIR1 , IL1 R2, IL1 RL1, IL1RL2, IL2RA, IL2RB, IL2RG, IL3RA, IL4R, IL5RA, IL6R, IL7R, IL8RA, IL8RB, IL9R, ILI0RA, ILI0RB, IL11RA, IL13RA, IL12RB2, , IL15RA, IL17R, IL18R1, IL20RA, IL21R, IL22R, IL1 HY1, IL1 RAP, IL1 RAPL1, IL1 RAPL2, IL1 RN, IL6ST, IL18BP, IL18RAP, IL22RA2, AIFI, HGF, LEP (leptin), PTN and THPO.

[0218] In another embodiment, a target molecule is a chemokine, chemokine receptor, or chemokine-related protein selected from the group consisting of CCLI (I-309), CCL2 (MCP -1 / MCAF), CCL3 (MIP- la), CCL4 (MIP-1b), CCL5 (RANTES), CCL7 (MCP-3), CCL8 (mcp-2), CCLH (eotaxin), CCL13 (MCP-4), CCL15 (MIP-1d), CCL16 ( HCC-4), CCL17 (TARC), CCL18 (PARC), CCL19 (MDP-3b), CCL20 (MIP-3a), CCL21 (SLC / exodus 2), CCL22 (MDC / STC-I), CCL23 (MPIF- I), CCL24 (MPIF-2 / eotaxin 2), CCL25 (TECK), CCL26 (eotaxin 3), CCL27 (CTACK / ILC), CCL28, CXCLI (GROI), CXCL2 (GRO2), CXCL3 (GR03), CXCL5 ( ENA-78), CXCL6 (GCP-2), CXCL9 (MIG), CXCL10 (IP 10), CXCL11 (1-TAC), CXCL12 (SDFI), CXCL13, CXCL14, CXCL16, PF4 (CXCL4), PPBP (CXCL7) , CX3CL1 (SCYDI), SCYEI, XCLI (lymphotactin), XCL2 (SCM-lb), BLRI (MDR15), CCBP2 (D6 / JAB61), CCR1 (CKRI / HM145), CCR2 (mcp-IRB / RA), CCR3 ( CKR3 / CMKBR3), CCR4, CCR5 (CMKBR5 / ChemR13), CCR6 (CMKBR6 / CKR-L3 / STRL22 / DRY6), CCR7 (CKR7 / EBII), CCR8 (CMKBR8 / TERI / CKR-LI), CCR9 (GPR-9-6), CCRLI (VSHKI), CCRL2 (L-CCR), XCRI (GPR5 / CCXCRI), CMKLRI, CMKORI (RDCI), CX3CR1 (V28), CXCR4, GPR2 (CCRIO), GPR31, GPR81 (FKSG80), CXCR3 (GPR9/CKR-L2), CXCR6 (TYMSTR /STRL33 / Bonzo), HM74, IL8RA (IL8Ra), IL8RB (IL8Rb), LTB4R (GPR16), TCPIO, CKLFSF2 , CKLFSF3, CKLFSF4, CKLFSF5, CKLFSF6, CKLFSF7, CKLFSF8, BDNF, C5R1, CSF3, GRCCIO (CIO), EPO, FY (DARC), GDF5, HDFIA, DL8, PRL, RGS3, RGS13, SDF2, SLIT2, TLR2, TLR2 , TREMI, TREM2 and VHL.

[0219] In another embodiment, heteromultima c proteins according to the present invention are capable of binding one or more targets selected from the group consisting of ABCFI; ACVRI; ACVRIB; ACVR2; ACVR2B; ACVRLI; ADORA2A; Aggrecan; AGR2; AICDA; AIFI; AIGI; AKAPI; AKAP2; AMH; AMHR2; ANGPTI; ANGPT2; ANGPTL3; ANGPTL4; ANPEP; APC; APOCI; AIR; AZGPI (zinc glycoprotein a); B7.1; B7.2; BAD; BAFF (BLys); BAGI; BAH; BCL2; BCL6; BDNF; BLNK; BLRI (MDR15); BMPI; BMP2; BMP3B (GDFIO); BMP4; BMP6; BMP8; BMPRIA; BMPRIB; BMPR2; BPAGI (plectin); BRCAI; C19orf10 (IL27w); C3; C4A; C5; C5R1; CANTI; CASP1; CASP4; CAVI; CCBP2 (D6 / JAB61); CCLI (1 -309); CCLII (eotaxin); CCL13 (MCP-4); CCL15 (MIP-Id); CCL16 (HCC-4); CCL17 (TARC); CCL18 (PARC); CCL19 (MIP-3b); CCL2 (MCP-1); MCAF; CCL20 (MIP-3a); CCL21 (MTP-2); SLC; exodus 2; CCL22 (MDC / STC-I); CCL23 (MPIF-1); CCL24 (MPIF-2 / eotaxin 2); CCL25 (TECK); CCL26 (eotaxin 3); CCL27 (CTACK / ILC); CCL28; CCL3 (MTP-1a); CCL4 (MDP-1b); CCL5 (RANTES); CCL7 (MCP-3); CCL8 (mcp-2); CCNAI; CCNA2; CCNDI; CCNEI; CCNE2; CCRI (CKRI / HM145); CCR2 (mcp-IRB / RA); CCR3 (CKR3 / CMKBR3); CCR4; CCR5 (CMKBR5 / ChemR13); CCR6 (CMKBR6 / CKR-L3 / STRL22 / DRY6); CCR7 (CKR7 / EBII); CCR8 (CMKBR8 / TERI / CKR-LI); CCR9 (GPR-9-6); CCRLI (VSHKI); CCRL2 (L-CCR); CD164; CD19; CDIC; CD20; CD200; CD22; CD24; CD28; CD3; CD37; CD38; CD3E; CD3G; CD3Z; CD4; CD40; CD40L; CD44; CD45RB; CD52; CD69; CD72; CD74; CD79A; CD79B; CD8; CD80; CD81; CD83; CD86; CDHI (E-cadherin); CDH10; CDH12; CDH13; CDH18; CDH19; CDH20; CDH5; CDH7; CDH8; CDH9; CDK2; CDK3; CDK4; CDK5; CDK6; CDK7; CDK9; CDKNIA (p21Wapl/Cipl); CDKNIB (p27Kipl); CDKNIC; CDKN2A (P16INK4a); CDKN2B; CDKN2C; CDKN3; CEBPB; CERI; CHGA; CHGB; Chitinase; CHST10; CKLFSF2; CKLFSF3; CKLFSF4; CKLFSF5; CKLFSF6; CKLFSF7; CKLFSF8; CLDN3; CLDN7 (claudin 7); CLN3; CLU (clusterin); CMKLRI; CMKORI (RDCI); CNRI; COL18A1; COLIAI; COL4A3; COL6A1; CR2; CRP; CSFI (M-CSF); CSF2 (GM-CSF); CSF3 (GCSF); CTLA4; CTNNBI (b-catenin); CTSB (cathepsin B); CX3CL1 (SCYDI); CX3CR1 (V28); CXCLI (GROI); CXCL10 (IP-10); CXCLII (1-TAC / IP-9); CXCL12 (SDFI); CXCL13; CXCL14; CXCL16; CXCL2 (GRO2); CXCL3 (GRO3); CXCL5 (ENA-78 / LIX); CXCL6 (GCP-2); CXCL9 (MIG); CXCR3 (GPR9/CKR-L2); CXCR4; CXCR6 (TYMSTR /STRL33 / Bonzo); CYB5; CYCI; CYSLTRI; DAB2IP; DES; DKFZp451 J01 18; DNCLI; DPP4; E2F1; ECGFI; EDGI; EFNAI; EFNA3; EFNB2; EGF; EGFR; ELAC2; ENG; ENO1; ENO2; ENO3; EPHB4; EPO; ERBB2 (Her-2); EREG; ERK8; ESRI; ESR2; F3 (TF); FADD; FasL; FASN; FCERIA; FCER2; FCGR3A; FGF; FGFI (aFGF); FGF10; FGF11; FGF12; FGF12B; FGF13; FGF14; FGF16; FGF17; FGF18; FGF19; FGF2 (bFGF); FGF20; FGF21; FGF22; FGF23; FGF3 (int-2); FGF4 (HST); FGF5; FGF6 (HST-2); FGF7 (KGF); FGF8; FGF9; FGFR3; FIGF (VEGFD); FELI (EPSILON); FILI (ZETA); FLJ12584; FLJ25530; FLRTI (fibronectin); FLTI; FOS; FOSLI (FRA-I); FY (DARC); GABRP (GABAa); GAGEBI; GAGECI; GALNAC4S-6ST; GATA3; GDF5; GFI1; GGT1; GM-CSF; GNASI; GNRHI; GPR2 (CCRIO); GPR31; GPR44; GPR81 (FKSG80); GRCCIO (CIO); GRP; GSN (Gelsolin); GSTPI; HAVCR2; HDAC4; HDAC5; HDAC7A; HDAC9; HGF; HYFIA; HDPI; histamine and histamine receptors; HLA-A; HLA-DRA; HM74; HMOXI; HUMCYT2A; ICEBERG; ICOSL; ID2; IFN-a; IFNAI; IFNA2; IFNA4; IFNA5; IFNA6; IFNA7; IFNB1; IFNgamma; DFNWI; IGBPI; IGFI; IGFIR; IGF2; IGFBP2; IGFBP3; IGFBP6; IL-I; IL10; IL10RA; IL10RB; IL11; IL11 RA; IL-12; IL12A; IL12B; IL12RB1; IL12RB2; IL13; IL13RA1; IL13RA2; IL14; IL15; IL15RA; IL16; IL17; IL17B; IL17C; IL17R; IL18; IL18BP; IL18R1; IL18RAP; IL19; IL1A; IL1B; ILIF10; IL1 F5; IL1 F6; IL1 F7; IL1 F8; IL1 F9; IL1 HYI; IL1 RI; IL1 R2; IL1 RAP; IL1 RAPL1; IL1 RAPL2; IL1 RL1; IL1 RL2, ILIRN; IL2; IL20; IL20RA; IL21R; IL22; IL22R; IL22RA2; IL23; IL24; IL25; IL26; IL27; IL28A; IL28B; IL29; IL2RA; IL2RB; IL2RG; IL3; IL30; IL3RA; IL4; IL4R; IL5; IL5RA; IL6; IL6R; IL6ST (glycoprotein 130); EL7; EL7R; EL8; IL8RA; DL8RB; IL8RB; DL9; DL9R; DLK; INHA; INHBA;INSL3; INSL4; IRAKI; ERAK2; ITGAI; ITGA2; ITGA3; ITGA6 (α6 integrin); ITGAV; ITGB3; ITGB4 (b 4 integrin); JAGI; JAKI; JAK3; JUN; K6HF; KAII; KDR; KITLG; KLF5 (GC Box BP); KLF6; KLKIO; KLK12; KLK13; KLK14; KLK15; KLK3; KLK4; KLK5; KLK6; KLK9; KRT1; KRT19 (Keratin 19); KRT2A; KHTHB6 (Hair-specific type H keratin); SLUDGES; LEP (leptin); Lingo-p75; Lingo-Troy; LPS; LTA (TNF-b); LTB; LTB4R (GPR16); LTB4R2; LTBR; MACMARKS; MAG or Omgp; MAP2K7 (c-Jun); MDK; MIBI; midkina; MEF; MIP-2; MKI67; (Ki-67); MMP2; MMP9; MS4A1; MSMB; MT3 (metallothionectin III); MTSSI; MUCI (mucin); MYC; MYD88; NCK2; neurocan; NFKBI; NFKB2; NGFB (NGF); NGFR; NgR-Lingo; NgR-Nogo66 (Nogo); NgR-p75; NgR-Troy; NMEI (NM23A); NOX5; NPPB; NROBI; NR0B2; NRIDI; NR1 D2; NR1 H2; NR1 H3; NR1 H4; NR1 I2; NR1 I3; NR2C1; NR2C2; NR2E1; NR2E3; NR2F1; NR2F2; NR2F6; NR3C1; NR3C2; NR4A1; NR4A2; NR4A3; NR5A1; NR5A2; NR6A1; NRPI; NRP2; NT5E; NTN4; ODZI; OPRDI; P2RX7; PAP; PARTY; PATE; PAWR; PCA3; PCNA; PDGFA; PDGFB; PECAMI; PF4 (CXCL4); PGF; PGR; phosphacan; PIAS2; PIK3CG; PLAU (uPA); PLG; PLXDCI; PPBP (CXCL7); PPID; PRI; PRKCQ; PRKDI; PRL; PROC; PROK2; PSAP; PSCA; PTAFR; PTEN; PTGS2 (COX-2); PTN; RAC2 (p21 Rac2); RARB; RGSI; RGS13; RGS3; RNFIIO (ZNF144); ROBO2; S100A2; SCGB1 D2 (lipophilin B); SCGB2A1 (mammaglobin 2); SCGB2A2 (managlobin 1 ); SCYEI (endothelial monocyte activating cytokine); SDF2; SERPINAI; SERPINA3; SERP1 NB5 (maspin); SERPINEI (PAI-I); SERPDMF1; SHBG; SLA2; SLC2A2; SLC33A1; SLC43A1; SLIT2; SPPI; SPRRIB(Sprl); ST6GAL1; STABI; STAT6; STEAP; STEAP2; TB4R2; TBX21; TCPIO; TDGFI; TEK; TGFA; TGFBI; TGFBIII; TGFB2; TGFB3; TGFBI; TGFBRI; TGFBR2; TGFBR3; THIL; THBSI (thrombospondin 1); THBS2; THBS4; THPO; TIE (Tie-1); TMP3; tissue factor; TLRIO; TLR2; TLR3; TLR4; TLR5; TLR6; TLR7; TLR8; TLR9; TNF; TNF-a; TNFAEP2 (B94); TNFAIP3; TNFRSPHIIA; TNFRSFIA; TNFRSFIB; TNFRSF21; TNFRSF5; TNFRSF6 (Fas); TNFRSF7; TNFRSF8; TNFRSF9; TNFSPHIO (TRAIL); TNFSFI 1 (TRANCE); TNFSF12 (APO3L); TNFSF13 (April); TNFSF13B; TNFSF14 (HVEM-L); TNFSF15 (VEGI); TNFSF18; TNFSF4 (OX40 linker); TNFSF5 (CD40 linker); TNFSF6 (FasL); TNFSF7 (CD27 linker); TNFSF8 (CD30 linker); TNFSF9 (4-1 BB linker); TOLLIP; Toll-like receptors; TOP2A (topoisomerase Ea); TP53; TPMI; TPM2; TRADD; TRAFI; TRAF2; TRAF3; TRAF4; TRAF5; TRAF6; TREMI; TRAIN2; TRPC6; TSLP; TWEAK; VEGF; VEGFB; VEGFC; versican; VHL C5; VLA-4; XCLI (lymphotactin); XCL2 (SCM-1b); XCRI(GPR5 / CCXCRI); YYI and ZFPM2.

[0220] Preferred molecular target molecules for antibodies encompassed by the present invention include CD proteins such as members CD3, CD4, CD8, CD16, CD19, CD20, CD34; CD64, CD200 of the ErbB receptor family, such as the EGF receptor, HER2, HER3 or HER4 receptor; cell adhesion molecules such as LFA-1, Mad, p150.95, VLA-4, ICAM-1, VCAM, alpha4/beta7 integrin and alphav/beta3 integrin, including their alpha or beta subunits (such as anti-CD11a antibodies, anti-CD18 or anti-CD11b); growth factors such as VEGF-A and VEGF-C; tissue factor (TF); alpha interferon (alphaIFN); TNFalpha, interleukin such as IL-1beta, IL-3, IL-4, IL-5, IL-8, IL-9, IL-13, IL-17A/F, IL-18, IL-13Ralpha1, IL- 13Ralpha2, IL-4R, IL-5R, IL-9R, IgE; blood group antigens; flk2/flt3 receiver; obesity receptor (OB); mpl receiver; CTLA-4; RANKL protein, RANK, RSV F, protein C etc.

[0221] In one embodiment, the heteromultimeric proteins according to the present invention bind protein relative to low density lipoprotein receptor (LRP) 1 or LRP 8 or transferrin receptor and at least one target selected from the group consisting of (1) beta-secretase (BACE1 or BACE2), (2) alpha-secretase, (3) gamma secretase, (4) taussecretase, (5) amyloid precursor protein (APP), (6) death receptor 6 (DR6), (7) beta amyloid peptide, (8) alphasynuclein, (9) Parkin, (10) Huntingtin, (11) p75 NTR and (12) caspase 6.

[0222] In one embodiment, the heteromultimeric proteins according to the present invention bind to at least two target molecules selected from the group consisting of: IL-1 alpha and IL-1 beta, IL-12 and IL- 18; IL-13 and IL-9; IL-13 and IL-4; IL-13 and IL-5; IL-5 and IL-4; IL-13 and IL-1 beta; IL-13 and IL-25; IL-13 and TARC; IL-13 and MDC; IL-13 and MEF; IL-13 and TGF-β; IL-13 and LHR agonist; IL-12 and TWEAK, IL-13 and CL25; IL-13 and SPRR2a; IL-13 and SPRR2b; IL-13 and ADAM8, IL-13 and PED2, IL17A and IL17F, CD3 and CD19, CD138 and CD20; CD138 and CD40; CD19 and CD20; CD20 and CD3; CD38 and CD138; CD38 and CD20; CD38 and CD40; CD40 and CD20; CD-8 and IL-6; CD20 and BR3, TNF alpha and TGF-beta, TNF alpha and IL-1 beta; TNF alpha and IL-2, TNF alpha and IL-3, TNF alpha and IL-4, TNF alpha and IL-5, TNF alpha and IL6, TNF alpha and IL8, TNF alpha and IL-9, TNF alpha and IL- 10, TNF alpha and IL-1 1 , TNF alpha and IL-12, TNF alpha and IL-13, TNF alpha and IL-14, TNF alpha and IL-15, TNF alpha and IL-16, TNF alpha and IL- 17, TNF alpha and IL-18, TNF alpha and IL-19, TNF alpha and IL-20, TNF alpha and IL-23, TNF alpha and IFNalpha, TNF alpha and CD4, TNF alpha and VEGF, TNF alpha and MIF, TNF alpha and ICAM-1 , TNF alpha and PGE4, TNF alpha and PEG2, TNF alpha and RANK ligand, TNF alpha and Te38; TNF alpha and BAFF; TNF alpha and CD22; TNF alpha and CTLA-4; TNF alpha and GP130; TNFa and IL-12p40; VEGF and HER2, VEGF-A and HER2, VEGF-A and PDGF, HER1 and HER2, VEGF-A and VEGF-C, VEGF-C and VEGF-D, HER2 and DR5, VEGF and IL-8, VEGF and MET, VEGFR and MET receptor, VEGFR and EGFR, HER2 and CD64, HER2 and CD3, HER2 and CD16, HER2 and HER3; EGFR(HERI ) and HER2, EGFR and HER3, EGFR and HER4, IL-13 and CD40L, IL4 and CD40L, TNFR1 and IL-1R, TNFR1 and IL-6R and TNFR1 and IL-18R, EpCAM and CD3, MAPG and CD28, EGFR and CD64, CSPGs and RGM A; CTLA-4 and BTNO2; IGF1 and IGF2; IGF1/2 and Erb2B; MAG and RGM A; NgR and RGM A; NogoA and RGM A; OMGp and RGM A; PDL-I and CTLA-4; and RGM A and RGM B.

[0223] Soluble antigens or their fragments, optionally conjugated to other molecules, can be used as immunogens for the generation of antibodies. For transmembrane molecules, such as receptors, their fragments (for example, the extracellular domain of a receptor) can be used as an immunogen. Alternatively, cells expressing the transmembrane molecule can be used as an immunogen. These cells can be derived from a natural source (such as cancer cell lines) or they can be cells that have been transformed by recombinant techniques to express the transmembrane molecule. Other antigens and their useful ways of preparing antibodies will be evident to those skilled in the art.

[0224] v. Embodiments: The present invention provides additional embodiments as described below. In a first embodiment, a method of producing heteromultimeric protein is provided, wherein said method comprises: obtaining a protein A-purified first hinge-containing polypeptide; obtaining a second protein A-purified hinge-containing polypeptide; adjusting the pH of each half-antibody from 4 to 9; mixing the first and second hinge-containing polypeptides to obtain a mixed hinge-containing polypeptide pool; adding a molar excess of a weak reductant to the mixed linkage-containing polypeptide pool; and incubating the hinge-containing polypeptide assembly mixed with the weak reductant to form a heteromultimeric protein comprising the first and second hinge-containing polypeptides.

[0225] In a second embodiment and according to the first embodiment, the first and second hinge-containing polypeptides can be selected from a half-antibody, immunoadhesin and its fragments. In a third embodiment and according to the first embodiment, the first hinge-containing polypeptide is a half-antibody. In a fourth embodiment and in accordance with the first embodiment, the second hinge-containing polypeptide is an Fc component. In a fifth embodiment and according to the third embodiment, the half-antibody comprises a VL domain, CL domain, VH domain, CH1 domain, hinge domain, CH2 domain, and CH3 domain. In a sixth embodiment and according to the fifth embodiment, the half-antibody is a single polypeptide chain which further comprises a tether in which said domains are positioned from each other in an N-terminal to C-terminal direction as follows: VL- CL-tether-VH-CH1-joint-CH2-CH3. In a seventh embodiment and according to the first embodiment, the first and second hinge-containing polypeptides are mixed prior to purification with protein A and co-purified on protein A. In an eighth embodiment and according to the first embodiment, the first and second embodiments hinge-containing polypeptides comprise a heteromultimerization domain. In a ninth embodiment and according to the eighth embodiment, the heteromultimerization domain is selected from a knob-in-hole mutation, leucine zipper, electrostatic etc. mutation. In a tenth embodiment and according to the ninth embodiment, the first hinge-containing polypeptide comprises a bulge (knob) and the second hinge-containing polypeptide comprises a hole (hole). In an eleventh embodiment and according to the first embodiment, the pH is adjusted after mixing. In a twelfth embodiment and according to the first or eleventh embodiment, the method further comprises adding L-arginine to a final concentration of 20 mM to 1 M before adjusting the pH. In a thirteenth embodiment and according to the first embodiment, the method further comprises incubating the mixed set at a temperature of 15 °C to 39 °C for at least thirty minutes. In a fourteenth embodiment and in accordance with the first embodiment, the assembly mixture has an oxidation potential of -200 to -600 mV, more preferably -300 to -500 mV, and preferably higher, about - 400 mV. In a fifteenth embodiment and according to the first embodiment, the weak reductant is selected from GSH, betamercaptoethylamine, cysteine/cysteine, GSH/GSSG, cysteamine/cystamine, glycylcysteine and betamercaptoethanol. In a sixteenth embodiment and in accordance with the first embodiment, the weak reductant is added in a 50-600 molar excess. In a seventeenth embodiment and in accordance with the first embodiment, the weak reductant is added before mixing. In an eighteenth run and according to the seventeenth run, the addition is carried out in less than one hour before mixing. In a tenth implementation and according to the first implementation, the assembly mixture incubation step is carried out at a temperature of 15 °C to 39 °C in the presence of polyvinylpyrrolidone (PVP). In a twentieth embodiment and according to the nineteenth embodiment, histidine is added before, simultaneously, or after PVP. In a 21st realization and according to the nineteenth realization, PVP is added up to 40% (w/v).

[0226] In a 22nd embodiment, the present invention provides a method of producing bispecific antibody, wherein said method comprises: a. obtaining a protein A-purified first half-antibody; B. obtaining a second half-antibody purified by protein A; ç. adding a solution of L-arginine to each half-antibody; d. adjusting the pH of each half-antibody from 4 to 9; and. mixing the first and second sets of half-antibodies to obtain a set of mixed half-antibodies; f. adding a molar excess of a weak reductant to the pool of mixed half-antibodies; and g. incubating the pool of mixed media-antibodies at a temperature of 15°C to 39°C in the presence of PVP; wherein a bispecific antibody comprising the first and second half-antibodies is produced.

[0227] In a 23rd embodiment, the present invention provides a method of producing heteromultimer, wherein said method comprises: (a) providing a mixture containing L-arginine of at least two different linkage containing polypeptides, wherein said mixture has a pH of 7 to 9; (b) addition of a molar excess of a weak reductant; and (c) incubating the mixture under conditions whereby a heteromultimer is produced.

[0228] In a 24th embodiment and according to the 22nd embodiment, the first half-antibody is a single-chain polypeptide comprising: (a) a full-length light chain comprising a VL domain and a CL domain; (b) a tether; and (c) a full length heavy chain comprising a VH domain, CH1 domain, hinge, CH2 domain and CH3 domain; wherein said polypeptide comprises light and heavy chain domains positioned together in an N-terminal to C-terminal direction as follows: V1-CL-CL/VH-teter-VH-CH1-hinge-CH2-CH3. In a 25th embodiment and according to the 24th embodiment, the single-chain polypeptide further comprises a heteromultimerization domain. In a 26th embodiment and according to the 25th embodiment, the heteromultimerization domain is a hole (such as a cavity-hole) or a bulge (knob). In a 27th embodiment and according to the 26th embodiment, the second half-antibody comprises a hole (hole) when the first half-antibody comprises a bulge (knob). In a 28th embodiment and according to the 26th embodiment, the second half-antibody comprises a knob when the first half-antibody comprises a hole (hole). In a 29th embodiment and according to the 24th embodiment, the tether comprises GGS repeats. In a 30th embodiment and according to the 24th embodiment, the teter contains from 15 to 50 amino acids.

[0229] In a 31st embodiment, the present invention provides a method of producing heteromultimeric proteins, wherein said method comprises: (a) obtaining a first hinge-containing polypeptide purified by protein A; (b) obtaining a second protein A-purified hinge-containing polypeptide; (c) pH adjustment of each hinge-containing polypeptide to 4 to 9 in the presence of L-arginine; (d) mixing the first and second hinge-containing polypeptides to obtain a mixed hinge-containing polypeptide pool; and incubating to form a heteromultimeric protein comprising the first and second hinge-containing polypeptides.

[0230] In a 32nd embodiment and according to the first or 31st embodiment, at least one of the half-antibodies is a single-chain polypeptide comprising: (a) a full-length light chain comprising a VL domain and a domain CL; (b) a tether; (c) a full length heavy chain comprising a VH domain, CH1 domain, hinge, CH2 domain and CH3 domain.

[0231] In a 33rd embodiment and according to the first embodiment, the first hinge-containing polypeptide is a single-chain polypeptide chain comprising light and heavy chain domains positioned from each other in the N-terminal to C-terminal direction as follows: VL-CL-VH-CH1-joint-CH2-CH3. In a 34th embodiment and according to the 33rd embodiment, the single-chain polypeptide chain further comprises a tether wherein said domains are positioned together in an N-terminal to C-terminal direction as follows: VL-CL-teter- VH-CH1-joint-CH2-CH3.

[0232] In a 35th embodiment, the present invention provides a method of producing heteromultimers, wherein said method comprises providing an L-arginine containing a mixture of hinge-containing polypeptides, wherein said mixture has a pH of 4 to 9, addition of a weak reductant and incubation under heteromultimer production conditions.

[0233] In a 36th embodiment, the present invention provides a host cell that has been designed to express a half-antibody, wherein said half-antibody is a single-chain polypeptide comprising a teter, VL domain, CL domain, domain VH, CH1 domain, hinge domain, CH2 domain and a CH3 domain, wherein said domains are positioned from each other in the N-terminal to C-terminal direction as follows: VL-CL-teter-VH-CH1-joint-CH2 -CH3. In a 37th embodiment and in accordance with the 36th embodiment, the single-chain polypeptide further comprises a heterodimerization domain. In a 38th embodiment and according to the 36th or 37th embodiment, the host cell is selected from prokaryotic cells, eukaryotic cells, mammalian cells or plant cells. In a 39th embodiment and according to the 38th embodiment, the host cell is a prokaryotic cell. In a 40th embodiment and according to the 39th embodiment, the prokaryotic cell is an E. coli cell. In a 41st embodiment and according to the 40th embodiment, the E. coli cell is lpp-deficient. In a 42nd embodiment and according to the 38th embodiment, the host cell is a mammalian cell. In a 43rd embodiment and in accordance with the 41st embodiment, the mammalian cell is a CHO cell. In a 44th embodiment and according to the 36th or 37th embodiment, the host cell comprises a vector encoding the single chain half antibody. In a 45th embodiment and according to the 36th or 37th embodiment, the single-chain polypeptide further comprises a heterodimerization domain. In a 46th embodiment and according to the 45th embodiment, the heterodimerization domain is selected from a knob, hole, one or more charged amino acids within the interface that are electrostatically unfavorable to homodimer formation but electrostatically favorable to heterodimer formation, one or more amino acids are altered to increase intramolecular ionic interactions, a supercoiled helix, and a leucine zipper.

[0234] In a 47th embodiment, the present invention provides a host cell mixture comprising a first host cell designed to express a first half-single-chain antibody and a second host cell designed to express an Fc component. In a 48th embodiment and according to the 36th embodiment, the half-antibody produced by a host cell comprises a heterodimerization domain.

[0235] In the experimental description that follows, the following abbreviations apply: eq (equivalents); M (molar); µM (micromolar); N (normal); mol (moles); mmol (millimoles); µmol (micromoles); nmol (nanomoles); g (grams); mg (milligrams); kg (kilograms); µg (micrograms); l (liters); ml (milliliters); µl (microliters); cm (centimeters); mm (millimeters); µm (micrometers); nm (nanometers); °C (degrees centigrade); h (hours); min (minutes); sec (seconds); msec (milliseconds). EXAMPLES

[0236] The present invention is described in further details in the examples below which are not intended in any way to limit the scope of the present invention as claimed. The attached figures are intended to be considered integral parts of the descriptive report and description of the present invention. All references mentioned are specifically incorporated herein by reference to everything described herein. EXAMPLE 1

[0237] Expression and purification: This example illustrates the expression and purification of half-antibodies.

[0238] Examples of methods of constructing and expressing half-antibodies in E. coli can be found, for example, in copending U.S. Application 2011/0287009, which is hereby incorporated in its entirety by reference. It is within the capacity of technicians in the field to modify and adjust the conditions of expression and cultivation.

[0239] Expression of half-antibodies in E. coli cells: Construction of expression plasmids: The heavy and light chain DNA coding sequences were cloned into an expression plasmid that contained separate promoter elements for each of the sequences and antibiotic resistance for selection of bacterial cells that contain the expression plasmid. The vector constructs also encode the heat stable enterotoxin II secretion signal (STII) ( Picken et al, 1983, Infect. Immun. 42: 269-275 and Lee et al, 1983, Infect. Immun. 42: 264- 268) for export of antibody polypeptides to the periplasmic space of the bacterial cell. Transcription of each strand is controlled by the phoA promoter (Kikuchi et al, 1981, Nucleic Acids Res., 9: 56715678) and translation control is provided by previously described STII signal sequence variants with measured relative translation resistance, which contain silent codon changes in the translation initiation region (TIR) (Simmons and Yansura, Nature Biotechnol., 14: 629-634 and Simmons et al, 2002, J. Immunol. Methods, 263: 133-147).

[0240] Each half-antibody contained a bulge (knob) or hole made in the heavy chain as described in U.S. Patent No. 7,642,228. Briefly, a CH3 knob mutant was first generated. A library of CH3 hole mutants was then created by randomizing residues 366, 368 and 407 that are found in the vicinity of the knob on the partner CH3 domain. In the examples below, the knob mutation was T366W and the hole contained mutations T366S, L368A, and Y407V in an IgG1 or IgG4 main chain. Equivalent mutations in other immunoglobulin isotypes can be made by those skilled in the art. Furthermore, those skilled in the art will readily appreciate that it is preferred that the two half-antibodies used for the bispecific are the same isotype.

[0241] Expression and purification: Half-antibodies containing either the knob or hole mutations were generated in separate cultures by expressing the heavy and light chain constructs in a bacterial host cell, such as E. coli. Expression plasmids were introduced into E. coli 33D3 (Ridgway et al (1999) 59(11):2718) or 64B4 (W3110 .DELTA.fhuA .DELTA.phoA ilvG+.DELTA.prc spr43H1 .DELTA.degP) host lines .DELTA.manA lacI.sup.q .DELTA.ompT) and the transformers were selected on LB plates containing carbenicillin. Transformers were then used to inoculate an initial LB culture containing carbenicillin and this was cultured overnight with shaking at 30°C. The initial culture was diluted 100X in C.R.A.P. phosphate limiting media (Simmons et al, 2002, J. Immunol. Methods, 263: 133-147) containing carbenicillin, and the culture was grown for 24 hours with shaking at 30°C. The cultures were centrifuged and the cell pellets were frozen until the beginning of antibody purification. The pellets were thawed and resuspended in an extraction buffer containing 25 mM Tris base adjusted to pH 7.5 with hydrochloric acid, 125 mM NaCl and 5 mM EDTA (Tris extraction buffer or TEB) with volume ratio and weight of 100 ml of TEB per five grams of cell pellet and extracted by breaking the cells using microfluids by passing the resuspended mixture through a Microfluidics Corporation model 110F microfluidizer (Newton, Mass.) three times. The bacterial cell extract was then clarified by centrifugation for twenty minutes at 15,000 Xg and the supernatant was collected and filtered through a 0.22 micron acetate filter prior to purification.

[0242] Each half-antibody was separately purified by Protein A affinity chromatography. Clarified cell extracts of the half-antibody bulge (knob) were loaded onto a 1 ml column of HiTrap MABSELECT SURE® from GE Healthcare (Piscataway NJ) at 2 ml/min. After loading, the column was washed with ten column volumes (CV) of 40 mM sodium citrate, pH 6.0, 125 mM sodium chloride and 5 mM EDTA followed by five column volumes of 20 mM citrate of sodium at pH 6.0 to facilitate capture by the cation exchange column. Affinity-captured half-antibodies were eluted with ten column volumes (CV) of 0.2 mM acetic acid (pH 2-3).

[0243] Expression of half-antibodies in CHO cells: Construction of expression plasmids: The heavy chain and light chain cDNAs were under the control of cytomegalovirus (CMV) immediate early gene promoter and enhancer. Each CMV transcription start site is followed by dividing donor and receptor sequences, which define introns that are removed from the final transcripts (Lucas et al, High-Level Production of Recombinant Proteins in CHO Cells Using a Dicistronic DHFR Intron Expression Vector, Nucl. Acid Res. (1996) 24: 1774-9). The enzyme glutamine synthetase (GS) has been used as a selection marker for the development of stable cell lines ( Sanders et al, Amplification and Cloning of the Chinese Hamster Glutamine Synthetase Gene, The EMBO J. (1984) 3:65-71 ) and was under the control of SV40 amplifier and early promoter.

[0244] Cell culture: CHO cells were cultured in a proprietary DMEM/F12 based medium in shake flask containers at 37 °C and 5% CO 2 . Cells were passed at a seeding density of 3 x 105/ml, every three to four days.

[0245] Stable transfection: CHO cells were transfected using lipofectamine 2000 CD according to the manufacturer's recommendations (Invitrogen, Carlsbad CA). Transfected cells were centrifuged and seeded in selective medium based on DMEM/F-12 (glutamine free) with various concentrations of methionine sulfoximine (MSX). About three weeks after seeding, individual colonies were taken into 96-well plates. Colonies taken were evaluated to determine antibody production by taking the supernatant for ELISA analysis. Superior clones were scaled and evaluated based on antibody titers, favorable metabolism (mainly lactate consumption) and acceptable product quality attributes.

[0246] Expression: Each half-antibody was expressed in CHO cells. Two liters of culture were cultivated and harvested.

[0247] Purification of half-antibodies: Each half-antibody was captured on a MABSELECT SURE® column. The column was then washed with four column volumes (CV) of the following buffers: an equilibration buffer consisting of 50 mM TRIS pH 8.0, 150 mM NaCl and a wash buffer consisting of 0.4 Potassium Phosphate M, pH 7.0. Each arm was eluted in 0.15 M sodium acetate at pH 2.9.

[0248] The above-described methods of expression and purification of half-antibodies are generally applicable to IgG of different isotypes. EXAMPLE 2

[0249] Solubilizer and pH maintenance: The following example details how incubation of media-antibodies under intermediate pH directed conformational change and increased assembly efficiency and how the addition of a solubilizer such as arginine and histidine reduced the intermediate pH-induced precipitation of half-antibodies.

[0250] Protein A clusters of half-antibodies are inherently unstable due to the exposed inner surface of the CH2 and CH3 domains that are likely to contain hydrophobic patches that normally lie on the non-solvent exposed surface of antibodies. In this way, when adjusted to a pH of more than 4, half-antibody protein A pools tend to precipitate. The present inventors found that, with minimal concentration of L-arginine present (in certain examples >50 mM), the half-antibody was stabilized and remained in solution by adjusting the pH. This addition of a solubilizer such as arginine kept the half-antibody in solution, reduced turbidity by pH adjustment and increased the yield of bispecific assembly. See Figure 3. Arginine also protected purified bispecific and bispecific half-antibodies against aggregation formation during freezing. Similar precipitation-reducing effect on histidine was also observed.

[0251] The protein recovered from the protein A columns in Example 1 was used as the starting material for this example.

[0252] L-arginine (1 M, pH 9) was added to the purified Protein A protein to a final concentration of 50 to 600 mM. The solution was then titrated to higher pH using 1.5 M Tris base, pH 11, as needed. The step of raising to mid-pH after acid elution from the Protein A column is called mid-pH maintenance.

[0253] Due to the knob and hole mutations in the CH3 domain, bispecific antibodies have different degrees of flexibility compared to standard antibodies. As a result of this unique flexibility and the exposed inner surfaces of the CH2 and CH3 domains, half-antibodies apparently underwent conformational changes upon pH adjustment. See Figure 2.

[0254] In this experiment, the buds underwent change from monomer to non-covalently bound homodimer when the pH was adjusted to a pH of more than 4. See Figure 2B. The holes underwent conformational change from a smaller hydrodynamic radius to a larger hydrodynamic radius based on the retention time of size exclusion chromatography. The change started to occur at pH 5 and pH > 7 took the change to completion. See Figure 2A.

[0255] Size Exclusion Chromatography (SEC) for determination of aggregation and oligomeric state of antibodies was performed by means of HPLC chromatography. Briefly, protein A purified antibodies were applied to a Tosoh TSK gel SW2000 column on an Agilent 1200 HPLC system. Protein (medium IgG1 antibody produced in E. coli) was eluted with 0.2 M K3PO4 and 0.25 M KCl at pH 6.2 at a flow rate of 0.5 ml/min. The eluted protein was quantified using UV absorption and peak area integration. See Figures 2A and B. This change may represent a folding intermediate that was induced by pH change due to the higher flexibility of the half-antibody, especially half-knob and hole (hole) antibodies with domain mutations CH3.

[0256] Maintenance of intermediate pH, however, can result in precipitation of half-antibodies. As shown in Figure 3A, the presence of arginine reduced the pH-induced turbidity of Protein A-purified IgG1 half-antibodies (knob) IgG1. of IgG1 half-antibodies produced by E. coli. The final concentration of arginine was about 50 mM when the pH was titrated to 5.5, about 200 mM at pH 7.5 and about 400 mM at pH 8.5 (see Figure 3A). The presence of arginine also increased the yield of knob-in-hole bispecific antibody assembly by 15% (data not shown).

[0257] Similarly, histidine was able to reduce the pH-induced turbidity due to precipitation. "Bump" IgG1 antibody medium (knob) was purified from E. coli homogenate on a MABSELECT SURE® column, resulting in a Protein A pool with a concentration of 12 g/l antibody medium. A quarter volume of arginine hydrochloride or histidine hydrochloride was added to a final additive concentration of 200 mM, with an equivalent volume of purified water added for the “none” control. The pH of the sample was increased using concentrated sodium hydroxide (50% w/v or 19.1 N NaOH solution) added in drops and the data points were recorded. The pH was measured using an Orion Ross 81-03 microprobe. The turbidity of the solution was measured using a Hach 2100 laboratory turbidity meter.

[0258] The data in Figure 3B demonstrated that arginine (200 mM) and histidine (200 mM) reduced the pH-induced precipitation in IgG1 isolated from E. coli. In summary, intermediate pH induced the change of conformation of half-antibodies in favor of bispecific assembly and a solubilizer added to the intermediate pH maintenance step reduced the precipitation induced by reduced pH. EXAMPLE 3

[0259] Reduction: The following example details how the use of a reducing condition reduces aggregation, which results in further formation of the desired heteromultimer, such as a bispecific antibody. Glutathione added to an assembly mixture, for example, creates a weakly reducing condition that is advantageous for bispecific knob-in-hole assembly. Other reducers in a similar class such as BMEA (betamercaptoethylamine) may have a similar effect.

[0260] Aggregation can occur during the assembly of half-antibodies of the bulge (knob) and hole (hole) to form bispecifics. Increasing glutathione levels minimizes the amount of aggregation during assembly. On the other hand, strong reducers such as DTT at high concentrations can sometimes increase aggregation. Without limitation to specific mechanisms, rather than permanently reducing disulfide bonds, glutathione apparently alters disulfides that act as a catalyst for proper disulfide formation. With glutathione assemblies, buffer exchange is not necessary to form the hinge region disulfides in the bispecific product of interest, as is necessary for reoxidation when using a strong reducer. The addition of a chemical reoxidant is not necessary when using a weak reducer such as glutathione.

[0261] Glutathione concentrations can be expressed in terms of molarity or in terms of a molar ratio with respect to the amount of polypeptides containing articulation or half-antibodies present in the assembly mixture. The use of target molar ratio of reducing controls to protein concentration in the assembly mixture prevents over-reduction or under-reduction as a result of varying protein concentrations.

[0262] In this example, glutathione was added to the mixed media-antibodies in 2- to 200-fold molar excess. Samples were incubated at room temperature for 46 hours. On RP HPLC (Reverse Phase High Performance Liquid Chromatography), all samples were diluted with 0.1% trifluoroacetic acid to a maximum concentration of 1.0 mg/ml. Protein concentration was determined by photometric measurements at 280 nm. Four samples of 0.1% trifluoroacetic acid were injected prior to sample analysis. This ensured that the column was completely balanced. Protein A-purified E. coli-produced IgG1 antibody media were applied to Poros R2/20 2.1 mm D x 20 mm L on an Agilent 1200 HPLC system. Protein was eluted with a linear gradient from 38 to 49% Buffer A to 0.09% trifluoroacetic acid and 80% acetonitrile (Buffer B) in twenty minutes at a flow rate of 0.5 ml/min. The eluted protein was quantified using UV absorption and peak area integration. As shown in Figure 4A, the level of bispecific formation increased with increasing glutathione:Ab molar ratio. See Figure 4A.

[0263] Size exclusion chromatography (SEC) was performed to determine the aggregation and oligomeric state of antibodies. Briefly, IgG1 half-antibodies produced by E. coli and purified by protein A were applied to a Tosoh TSKgel SW2000 column on an Agilent 1200 HPLC system. Protein was eluted with 0.2 M K3PO4 and 0.25 M KCl at pH 6.2 at a flow rate of 0.5 ml/min. The eluted protein was quantified using UV absorption and peak area integration. The 150 kD peak observed was confirmed to be due to bispecific antibody formation. See Figure 4B.

[0264] As can be seen, there is a change in peaks of unwanted monomers (that is, half-antibodies, either from a knob or a hole (hole)) and homodimers to the heteromultimer, that is, a bispecific antibody (Figures 4A and B). In summary, the data demonstrate that increasing the molar ratio of glutathione to half-antibodies reduced aggregation and increased bispecific formation (Figure 4B). EXAMPLE 4

[0265] Temperature: This example illustrates the effect of temperature on the stability of half-antibodies and heteromultimer assembly.

[0266] The temperature of the half-antibody solution had a dramatic impact on the assembly speed. An example of increased assembly of IgG1 half-antibodies produced by E. coli under higher temperature is shown in Figure 5A.

[0267] Another example showing the effect of temperature on bispecific assembly is shown in Figure 5B. In this experiment, two half-IgG1 antibodies were produced in E. coli and purified on Protein A as described in Example 1. The half-antibodies were combined and divided into four plots for bispecific assembly testing under different conditions with or without heating and /or maintenance of intermediate pH.

[0268] As shown in Figure 5B, 200 molar excess of glutathione under varying conditions increased the rate of bispecific IgG1 antibody formation. The control conditions (room temperature, about 20 °C, the half-antibody was kept at pH 4, without maintenance of intermediate pH) allowed the bispecific assembly, although at a slower speed. Maintaining Protein A-purified half-antibodies at intermediate pH (pH 5 for half-antibody of knob; pH 7 for half-antibody of hole (hole)) for 16 hours at room temperature increased the rate of formation of bispecific antibodies without moving to higher temperature (assembly mixture had pH 8.5 at “optimized pH” in Fig. 5B). Increasing the temperature to 37 °C without an intermediate pH maintenance step increased the speed of assembly of bispecific antibody on the control and was faster relative to a pH maintenance step and assembly performed at room temperature. The highest assembly speed, however, was observed when the Protein A-purified half-antibodies were kept at intermediate pH (as above) and then assembled at pH 8.5 under elevated temperature (ie, 37 °C). Under this condition, about 80% bispecific assembly was achieved in just about six hours (Figure 5B). In summary, overall assembly speed was increased through heating, and maintaining pH and heating had a synergistic effect on assembly.

[0269] Heat-amplified mounting has also been observed in IgG4 bispecific antibody. The results of Figure 6B demonstrate that, in the presence of PVP and histidine, heated IgG4 half-antibodies produced by E. coli cultivation achieved assembly results similar to the assembly of IgG1 half-antibodies produced by E. coli. The amount of bispecific was analyzed using reverse phase HPLC as described above. Taken together, the data demonstrate that warming facilitated the formation of bispecifics. EXAMPLE 5

[0270] Stabilizers: The following example details how stabilizers can reduce aggregation as a result of heating and/or high pH during assembly and/or maintenance of intermediate pH.

[0271] Polyvinylpyrrolidone (PVP) is a water-soluble uncharged polymer with a pyrrolidone group. PVP reduced aggregation during heated assembly. Without limitation to specific mechanisms, PVP may act to stabilize a bispecific folding intermediate or protect half-antibodies against aggregation likely by interacting with bispecific hydrophobic patches.

[0272] The effect of PVP on the formation of aggregates was analyzed using SEC under the conditions described in Example 3. The addition of PVP minimized the high molecular weight substances (HMWS) present in the assembled set to 12% HMWS with PVP at 4 % (w/v) compared to 4% NMWS without the addition of PVP. See Figure 6A. All samples were IgG1 produced by E. coli heated in the presence of 200 mM arginine.

[0273] Next, the IgG4 bispecific antibody assembly was tested. As shown in Figure 6B, heated assembly in the presence of PVP and histidine greatly increased the assembly of bispecific IgG4 produced by E. coli to levels similar to the heated assembly of IgG1 produced by E. coli shown in Figure 5A. Arginine was present in the heated sample and in the sample at room temperature as a solubilizer and pH titrator. In addition to PVP, another histidine stabilizer was added prior to the heated intermediate pH maintenance step to stabilize half-antibody during this step. The results demonstrate that PVP and histidine increased IgG4 bispecific assembly to levels similar to IgG1 bispecific assembly (compare Figures 6B and 5B).

[0274] Figure 6C presents another example in which PVP minimized the formation of HMWS during the heated assembly of an IgG4 bispecific antibody.

[0275] Heating the sets of Protein A and half-antibodies accelerated the conformational changes of half-antibodies by incubation under intermediate pH. Heating can, however, cause aggregation, especially for the assembly of bispecific IgG4 bump and hole antibodies from half-IgG4 antibodies produced in E. coli. In this way, solubilizer and/or additional stabilizers were added by heating the IgG4 antibody media during assembly.

[0276] Conformation alteration of E. coli IgG4 half-hole antibody was detected after 48 hours of incubation at room temperature. When IgG4 half-hole antibodies were incubated at 37°C, conformational change was detected within about three hours (data not shown). Warming, however, generated increased aggregation as determined by the SEC. See Figure 7, left panel. In this experiment, histidine was added during the warm intermediate pH maintenance step to test its effect on reducing half-antibody aggregation. As shown in Figure 7, the presence of histidine during heating minimized the aggregation level from 11% high molecular weight substances (HMWS, no histidine) to 6% HMWS (200 mM histidine), without affecting the change in conformation of half-antibodies. The results therefore demonstrate that a stabilizer reduced the formation of aggregates during the assembly of bispecific antibodies, as well as the maintenance of intermediate pH of half-antibodies. EXAMPLE 6

[0277] Assembly: This example provides protocols for two examples of immunoglobulin isotypes, namely IgG1 and IgG4. The half-antibodies were produced in one of two different host cells, namely, E. coli or CHO. It is understood that the methods described herein can be applied to other isotypes of antibodies produced from the same or other sources. It is well within the skill of those skilled in the art to modify the protocols through routine experimentation based on knowledge of the art and the teachings described in the present application.

[0278] It is further understood that the formation of an antibody comprising a half-antibody produced in a first host cell (such as CHO) can be assembled with a complementary half-antibody produced in a second host cell (such as E. coli) (data not displayed). In this way, for example, a half-antibody produced in CHO can be assembled with a half-antibody produced in E. coli or vice versa.

[0279] All four assembly procedures described in this example resulted in assemblies that stabilized after four hours and produced less than 10% aggregate and minimal aggregation during assembly.

[0280] a. E. coli IgG1: Conformation change of half-antibodies: If the protein A sets had a pH of less than 7, the pH of the two sets of half-antibodies was adjusted to pH 7 using 1 M arginine (pH 9) and the two sets were incubated at 37 °C for three hours. Alternatively, sets were incubated at room temperature for 48 hours. If the sets have been at pH 7 for 48 hours or more, skip this step. The amount of arginine added to bring the solution to pH 7 was determined.

[0281] The sets (still warm) were combined and the pH was adjusted to pH 8.5 using 1 M arginine (pH 9). The amount of arginine added in this step was determined.

[0282] 2 M arginine (pH 8.5) was added until the final concentration of arginine was 0.5 M.

[0283] 200 mM reduced glutathione (GSH) in 0.5 M arginine (final pH 8.5) was added until the GSH:Ab ratio was 200X (eg add 6.88 µl of the glutathione solution for each mg of half-antibody).

[0284] The glutathione half-antibody solution was incubated at 37 °C for four hours to allow the assembly of the half-antibodies into a bispecific antibody.

[0285] b. CHO IgG1: Assembled as described above for E. coli. The pH may not need to be raised to pH 7 or above to maintain an intermediate pH. The assembly time after addition of glutathione may reach completion after two hours.

[0286] c. CHO IgG4: Assembled with the same conditions as above (CHO IgG1). Histidine and PVP may not be needed.

[0287] d. IgG4 from E. coli: The assembly of IgG4 from E. coli using the above protocols resulted in high levels of aggregate, ie around 35%. This necessitated modification of mounting conditions.

[0288] It was determined that the Protein A pool composition containing 0.2 M histidine and 50 mM arginine gave acceptable results (data not shown). In this way, several ways of providing the final result were researched and determined to provide acceptable results.

[0289] A first method used altered elution (pH 3) and wash buffers (pH 7) during Protein A purification to contain 0.2 M His and 50 mM Arg. This resulted in the final Protein A pool containing 0.2 M His and 50 mM Arg, pH 4, which was then titrated to pH 8 using 1.5 M Tris base (pH 11).

[0290] A second alternative method used a 0.8 M solution of Histidine HCl (which has a solubility of 0.8 M). One third volume of Histidine HCl was added to the Protein A pool(s) to achieve a final concentration of 0.2 M His. Then 1/40 volume of 2M Arg was added.

[0291] A third alternative method was to buffer exchange the Protein A pools to a buffer of 0.2 M His and 50 mM Arg (preferably under pH 8).

[0292] A final alternative method was the addition of histidine directly to the Protein A pool(s) (31.03 g/l) and then addition of 1/40 of the 2 M volume of Arg.

[0293] A quarter volume of a 20% w/v Spectrum PVP K-15 solution in 0.2 M histidine and 50 mM Arg was added to the Protein A pool(s).

[0294] It was observed that PVP also minimized aggregation during assembly for IgG1 under conditions of low glutathione content.

[0295] Change of conformation of half-antibodies: The pH of the protein A pool(s) was adjusted to pH 8.0 using 1.5 M Tris base with 0.2 M histidine and 50 mM of 4% Arg and PVP K-15 (pH 11) and incubated at 37 °C for three hours. Alternatively, the set(s) may be incubated at room temperature for 48 hours.

[0296] 200 mM reduced glutathione (GSH) in 0.2 M histidine, 4% PVP, 50 mM arginine (final pH 8.0) was added until the GSH:Ab ratio was 200X (ex. : add 6.88 μl of glutathione solution for each mg of antibody medium). If the half-antibody sets were not matched, they were matched at that point.

[0297] The pooled half-antibodies were incubated at 37 °C for four hours to allow the formation of the bispecific antibody. At that moment, the percentage of bispecific antibody stabilized.

[0298] After leveling off the amount of bispecific antibody, the solution can be stored at low temperature or adjusted to a lower pH for processing in subsequent chromatography steps.

[0299] The methods described herein find use in the manufacture of therapeutic proteins such as bispecific antibodies.

[0300] It is understood that the examples and realizations described herein are for illustrative purposes only and that various modifications or changes in their light will be suggested to those skilled in the art and should be included within the spirit and purpose of this application and the scope of the appended claims. All publications, patents, and patent applications referenced herein are incorporated in their entirety by reference for all purposes.

Claims (73)

1. METHOD OF PRODUCTION OF HETEROMULTIMERIC PROTEIN, characterized in that it comprises: (a) obtaining a first half-antibody, purified by protein A, at a pH of 5 to 9 in the presence of a first solubilizer, wherein the first half-antibody comprises a heterodimerization domain; (b) providing a second half-antibody, purified by protein A, at a pH of 5 to 9 in the presence of a second solubilizer, wherein the second half-antibody comprises a heterodimerization domain; (c) mixing the first and second half-antibodies to form an assembly mixture in a reducing condition comprising 50 to 400x molar excess of glutathione (GSH) relative to the total amount of half-antibody; and (d) incubating the assembly mixture to produce a heteromultimeric protein comprising the first and second half-antibodies, wherein the first half-antibody interacts with the second half-antibody in a heteromultimerization domain. 2. METHOD, according to claim 1, characterized in that the first or second solubilizer is selected from the group consisting of arginine, histidine and sucrose. 3. METHOD, according to claim 1, characterized in that the first or second solubilizer is arginine, an arginine salt or an arginine derivative. 4. METHOD, according to claim 1, characterized in that the first or second solubilizer is histidine, a histidine salt or a histidine derivative. 5. METHOD according to claim 2, characterized in that the first or second solubilizer is present at a concentration between 20 mM and 1 M. 6. METHOD, according to claim 5, characterized in that the first or second solubilizer is present in a concentration between 20 mM and 400 mM. 7. METHOD, according to claim 1, characterized in that the first and second sulubilizers are arginine HCl. 8. METHOD, according to claim 1, characterized in that the first half-antibody and second half-antibody are purified before mixing. 9. METHOD, according to claim 1, characterized in that the first half-antibody and the second half-antibody are co-purified. 10. METHOD, according to claim 1, characterized in that step (a) is preceded by the step of purification of the first half-antibody or step (b) is preceded by the step of purification of the second half-antibody. 11. METHOD, according to claim 10, characterized in that the first and second half-antibodies are produced by a bacterial cell, a yeast cell, a bacillovirus or a mammalian cell. 12. METHOD, according to claim 11, characterized in that the first and second half-antibodies are produced by a mammalian cell. 13. METHOD according to claim 12, characterized in that the mammalian cell is a CHO cell. 14. METHOD, according to claim 1, characterized in that the first or second half-antibody is an IgG half-antibody. 15. METHOD, according to claim 14, characterized in that the IgG half-antibody is of the IgG1, IgG2 or IgG4 isotype. 16. METHOD according to claim 14, characterized in that the first or second half-antibody comprises an Fc component. 17. METHOD according to claim 14, characterized in that the first or second half-antibody comprises a VL domain, a VH domain, a hinge domain, a CH2 domain and a CH3 domain. 18. The METHOD according to claim 17, characterized in that the first or second half-antibody comprises a single-chain polypeptide which additionally comprises a tether, and wherein said single-chain polypeptide comprises domains positioned relative to each other in an N-direction. terminal to C-terminal as follows: VL-CL-teter-VH-CH1-hinge-CH2-CH3. 19. METHOD according to claim 17, characterized in that the first or second half-antibody additionally comprises a CL domain and a CH1 domain. 20. The method according to claim 19, characterized in that the first or second half-antibody comprises a single-chain polypeptide which additionally comprises a tether, and wherein said single-chain polypeptide comprises domains positioned relative to each other in an N- direction. terminal to C-terminal as follows: VL-CL-teter-VH-CH1-hinge-CH2-CH3. 21. METHOD according to claim 1, characterized in that one or more of the steps (a)-(d) are heated to a temperature between 25 °C and 42 °C. 22. METHOD according to claim 21, characterized in that one or more of steps (a) to (d) are heated to a temperature between 32 °C and 37 °C. 23. METHOD, according to claim 21, characterized in that the first half-antibody of step (a) and the second half-antibody of step (b) are heated. 24. METHOD, according to claim 21, characterized in that the assembly mixture of step (d) is heated. 25. METHOD according to claim 21, characterized in that all steps (a)-(d) are heated. 26. METHOD according to claim 25, characterized in that one or more of steps (a) to (d) are heated to a temperature between 32 °C and 37 °C. 27. METHOD, according to claim 1, characterized in that the reduction condition has an oxidation potential between -200 to -600 mV. 28. METHOD according to claim 1, characterized in that GSH is added in molar excess of 100-300X to the assembly mixture. 29. METHOD according to claim 28, characterized in that GSH is added in molar excess of 200X to the assembly mixture. 30. METHOD according to claim 1, characterized in that the assembly mixture is incubated at a pH of 7 to 9. 31. The METHOD according to claim 1, characterized in that the heteromultimerization domain of the first or second half-antibody comprises one or more of a bulge, a hole (such as a cavity), leucine zipper, supercoiled helix or an amino acid residue polar capable of forming an electrostatic interaction. 32. METHOD, according to claim 31, characterized in that the heteromultimerization domain of the first half-antibody comprises a bulge and the heteromultimerization domain of the second half-antibody comprises a hole. 33. METHOD according to claim 1, characterized in that it additionally comprises the addition of a stabilizer in one or more of steps (a) to (d). 34. METHOD, according to claim 33, characterized in that the stabilizer is added in step (c) or in step (d). 35. METHOD, according to claim 34, characterized in that the stabilizer is polyvinylpyrrolidone (PVP). 36. METHOD, according to claim 1, characterized in that it further comprises the step of recovering the heteromultimeric protein. 37. METHOD according to claim 36, characterized in that the step of recovering the heteromultimeric protein comprises the purification of the heteromultimeric protein. 38. METHOD OF PRODUCTION OF BISPECIFIC ANTIBODY, characterized in that it comprises the steps of: (a) providing a first half-antibody at a pH of 5 to 9 in the presence of arginine or histidine, wherein the first half-antibody comprises a domain of heteromultimerization; (b) providing a second half-antibody at a pH of 5 to 9 in the presence of arginine or histidine, wherein the second half-antibody comprises a heteromultimerization domain; (c) mixing the first and second half-antibodies to form an assembly mixture under reducing condition comprising a 50-400x molar excess of glutathione (GSH) relative to the total amount of half-antibody; and (d) incubating the assembly mixture to form a bispecific antibody comprising the first and second half-antibodies, wherein the first half-antibody interacts with the second half-antibody in the heteromultimerization domain. 39. METHOD according to claim 38, characterized in that arginine or histidine is present at a concentration between 20 mM and 400 mM. 40. METHOD, according to claim 38, characterized in that the first half-antibody and the second half-antibody are purified before mixing. 41. METHOD according to claim 38, characterized in that the first and second half-antibodies are produced by a bacterial cell, a yeast cell, a bacillovirus or a mammalian cell. 42. METHOD according to claim 41, characterized in that the first and second half-antibodies are produced by a mammalian cell. 43. METHOD according to claim 42, characterized in that the mammalian cell is a CHO cell. 44. METHOD according to claim 38, characterized in that the first or second half-antibody is of the IgG1, IgG2 or IgG4 isotype. 45. The method according to claim 38, characterized in that the first or second half-antibody comprises an Fc component. 46. METHOD according to claim 38, characterized in that one or more of steps (a) to (d) are heated to a temperature between 25 °C and 42 °C. 47. METHOD according to claim 38, characterized in that GSH is added in molar excess of 100-300X to the assembly mixture. 48. The METHOD according to claim 38, characterized in that the heteromultimerization domain of the first or second half-antibody comprises one or more of a bulge, a hole (such as a cavity), leucine zipper, supercoiled helix or a polar amino acid residue capable of forming an electrostatic interaction. 49. The METHOD according to claim 38, characterized in that the first heterodimerization domain of the first half-antibody comprises a bulge and the heterodimerization domain of the second half-antibody comprises a hole. 50. METHOD according to claim 38, characterized in that it further comprises the addition of a stabilizer in one or more of steps (a) to (d). 51. METHOD, according to claim 50, characterized in that the stabilizer is added in step (c) or step (d). 52. METHOD, according to claim 51, characterized in that the stabilizer is arginine or polyvinylpyrrolidone (PVP). 53. METHOD, according to claim 38, characterized in that it additionally comprises the step of recovering the bispecific antibody. 54. METHOD, according to claim 53, characterized in that the step of recovering the bispecific antibody comprises the purification of the bispecific antibody. 55. METHOD OF PRODUCING HETEROMULTIMERIC PROTEIN, characterized in that it comprises: (a) obtaining a first half-antibody purified by protein A, wherein the first half-antibody comprises a heterodimerization domain; (b) obtaining a second half-antibody purified by protein A, wherein the second half-antibody comprises a heterodimerization domain; (c) adjust the pH of each half-antibody joint containing a polypeptide to between pH 4 and 9; (d) mixing the first and second half-antibodies to obtain an assembly mixture; (e) add 50 to 400x molar excess of glutathione (GSH) to the assembly mixture; and (f) incubating the assembly mixture to form a heteromultimeric protein comprising the first and second half-antibodies. 56. The method according to claim 55, characterized in that the first or second half-antibody comprises an Fc portion. 57. METHOD according to claim 55, characterized in that the pH of the first and second half-antibody is adjusted to pH 4 to 9 in the presence of a solubilizer. 58. METHOD according to claim 55, characterized in that the solubilizer is arginine or histidine that is added to a final concentration between 20 mM and 1 M before adjusting the pH. 59. The METHOD according to claim 55, characterized in that the heteromultimerization domain of the first or second half-antibody comprises one or more of a bulge, a hole (such as a cavity), leucine zipper, supercoiled helix or an amino acid residue polar capable of forming an electrostatic interaction. 60. The METHOD according to claim 59, characterized in that the heteromultimerization domain of the first half-antibody comprises a bulge and the heteromultimerization domain of the second half-antibody comprises a hole. 61. METHOD according to claim 55, characterized in that the pH is adjusted after mixing. 62. METHOD according to claim 55, characterized in that it further comprises incubating the assembly mixture at a temperature between 15 °C and 39 °C for at least 30 minutes. 63. METHOD according to claim 55, characterized in that the assembly mixture in step (f) has an oxidation potential between -200 to -600 mV. 64. METHOD, according to claim 55, characterized in that the incubation of the assembly mixture is carried out at a temperature between 15 °C and 39 °C in the presence of polyvinylpyrrolidone (PVP). 65. METHOD, according to claim 64, characterized in that PVP is added up to 40% (w/v). 66. METHOD according to claim 55, characterized in that the first and second half-antibodies are produced by a bacterial cell, a yeast cell, a bacillovirus or a mammalian cell. 67. METHOD according to claim 66, characterized in that the first or second half-antibody is produced by a CHO cell. 68. METHOD according to claim 55, characterized in that GSH is added in molar excess of 100-300X. 69. METHOD, according to claim 55, characterized in that GSH is added in molar excess of 200X. 70. METHOD, according to claim 60, characterized in that the protuberance comprises a replacement T366W and the orifice comprises replacements T366S, L368A and Y407V. 71. METHOD according to claim 55, characterized in that the pH is reduced after the heteromultimeric protein is formed in step (f). 72. METHOD according to claim 55, characterized in that the temperature is decreased after the heteromultimeric protein is formed in step (f). 73. METHOD OF PRODUCTION OF HETEROMULTIMERIC PROTEIN, characterized in that it comprises: (a) obtaining a first half-antibody, wherein the first half-antibody comprises a heterodimerization domain; (b) obtaining a second half-antibody, wherein the second half-antibody comprises a heterodimerization domain; (c) adjust the pH of each half-antibody to between pH 4 and 9; (d) mixing the first and second half-antibodies to obtain an assembly mixture; (e) adding a 50-400x molar excess of glutathione (GSH) to the assembly mixture; and (f) incubating the assembly mixture to form a heteromultimeric protein comprising the first half-antibody and the second half-antibody.

Images